Do Hooded crows (Corvus cornix) understand how a loose string works? 

Do Hooded crows (Corvus cornix) understand how a loose string works? | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Do Hooded crows (Corvus cornix) understand how a loose string works? Anna A. Smirnova, Maria A. Cheplakova, Ksenia N. Kubenko This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5405438/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 04 May, 2025 Read the published version in Scientific Reports → Version 1 posted 11 You are reading this latest preprint version Abstract In a loose-string task an out-of-reach tray baited with food can only be retrieved by simultaneously pulling on both ends of a string threaded through the loops on the tray. This task is used to assess an animal's ability to cooperate, with each animal only having access to one end of the string. Some studies use the loose-string task in a pre-training phase, during which animals are individually taught to pull both ends of the string. Usually, no additional tests are conducted to determine whether the animals have understood how the loose string works. It is conceivable that a lack of knowledge of the causal basis of the loose-string task could make it more challenging to grasp how the partner can assist with it. Here, we tested whether Hooded crows could acquire some knowledge of the causal basis of the loose-string task. Prior to the critical test (Experiment 3), the birds were presented with two different tasks (Experiment 1 and 2) to allow them to acquire some knowledge of the causal basis of the task. The results may indicate that, as a consequence of the experience gained, some crows may have begun to understand how the loose string works. Biological sciences/Neuroscience/Cognitive neuroscience/Problem solving Biological sciences/Neuroscience/Cognitive neuroscience/Intelligence physical cognition problem-solving causal reasoning cooperation loose-string paradigm corvids Figures Figure 1 Figure 2 Introduction In the domain of physical cognition, one of the most widely used paradigms is the string-pulling task (e.g. 1 , 2 , 3 , 4 , 5 , 6 ), in which the animal must pull the baited string in order to obtain the bait. Even when using a series of such tasks of varying complexity, it's challenging to distinguish between solutions produced by reinforcement of random behaviour (i.e. induced by associative learning) and those resulting from understanding or reasoning 3 , 7 . Bluff and colleagues 8 were correct to point out that it is challenging to distinguish between associatively learned behaviour and associatively learned knowledge. A variation of the string-pulling task can be considered a loose-string task, in which an out-of-reach tray baited with food rewards can be retrieved only by simultaneously pulling on both ends of a string threaded through the loops on the tray. Thе task, in which each of the two animals only has access to one end of the string, is commonly used for studies on animal cooperation. Some studies use the loose-string task in a pre-training phase, during which animals are individually taught to pull both ends of the string simultaneously 9 , 10 , 11 , 12 , 13 , 14 , 15 , 16 , 17 . Usually, no additional tests are conducted to determine whether the animals have understood how the loose string works. We found a single study that conducted a transfer test at the end of the training phase to assess whether the animals could generalise from the training trials to complete a task that presented the string in a novel manner 13 . Understanding of the causal basis of the loose-string task has rarely but been mentioned (e.g. 11 , 18 ) when discussing factors influencing the outcome of a cooperative task. Although there is another point of view that the ability of the animals to learn the role of their partners neither implies nor requires that they understood how the loose string and the apparatus itself worked (e.g. 19 ). It is conceivable that a lack of knowledge of the causal basis of the loose-string task could make it more challenging to understand how the partner can assist with it. In the current study, we investigated whether Hooded crows ( Corvus cornix ) could acquire some knowledge of the causal basis of the loose-string task. Сrows possess advanced cognitive abilities. They can extract relations among items and between relations, form abstract categories not tied to specific perceptual features and use abstract representations 20 , 21 , 22 , 23 , 24 , 25 , 26 , 27 , 28 , 29 , 30 , 31 . The advanced cognitive abilities of corvids are determined by the high level of their brain complexity 32 , 33 , 34 . They are characterised by an expanded associative meso- and nidopallium 35 , 36 . Meso- and nidopallium of corvids (at least in Carrion crows) is more densely and diversely innervated by dopaminergic fibres and they have more expended nidopallium caudolaterale — the functional analogue to the mammalian prefrontal cortex 37 . Hooded crows are not specialised tool users, but like other members of the corvid family 38 , 39 , they drop shells on rocks or nuts on motorways, which could be considered an example of proto-tool use. They throw twigs at intruders during the nesting season, although this may be a manifestation of displacement behaviour, which is typical for them when they are frustrated 40 . As previously demonstrated by Bagotskaya et al. (2012), some Hooded crows are capable of solving sophisticated variants of horizontal string-pulling tasks for which the proximity rule is not applicable. Hauever, it remains unclear whether crows understand the principles underlying a string-pulling task or whether this result is due to rapid associative learning by using perceptual-motor feedback 2 , 3 . We have recently found 41 that Hooded crows similar to tool-specialised New Caledonian crows 42 and non-tool-specialised Goffin's cockatoos 43 , can manufacture objects according to a mental template. In the present study, a set of three tasks was applied, comprising one standard task (Experiment 2) and two novel ones. Prior to the critical test (Experiment 3), the birds were presented with two different loose-string tasks (Experiment 1 and Experiment 2) to allow them to acquire some knowledge of the causal basis of the task. In the initial two experiments, the ability of the crows to cope with the task spontaneously was assessed, and subsequently, the birds were trained to solve it if they failed. The third task was designed to assess whether the experience gained during the previous two tasks had resulted in the acquisition of some knowledge about the causal basis of the loose-string task. Methods Subjects. Six adult Hooded crows ( Corvus cornix ) were tested. All of them were rescued from the wild due to injuries and were housed in the outdoor group aviary (500 × 250 × 300 cm) on the territory of the Lomonosov Moscow State University Botanical Garden. The aviaries are equipped with perches (tree branches, wooden ladders, stumps), wooden shelters, toys, metal and ceramic feeders and plastic basins with water. The birds are kept on an ad libitum diet (rat and mouse carcasses, steamed crops and seeds with added vegetable oil and vitamins, eggs, seasonal fruits and vegetables, and fresh water). If the crows refused to participate in the experiment, then they received food without animal protein for 1 or 2 days. Glaz and Schnobel were kept for over 15 years, and Rodya, Joe, Grisha, and Clara for over four. The sex of the birds is not known exactly. Based on sex-size dimorphism and behaviour, it can be assumed that Grisha and Clara are females, and other crows are males. Prior to this experiment, all birds, except Grisha, had participated in experiments based on Aesop's fable paradigm. Two of the subjects (Glaz and Schnobel) had previously participated in a series of highly varied identity matching-to-sample tasks and later spontaneously performed relational matching-to-sample tasks 26 . Glaz and Schnobel participated in a string-pulling experiment 44 . None of the other subjects had any prior experience with strings. The experiments were conducted from 2021 to 2023. All experiments were appetitive, non-invasive and based exclusively on behavioural tests. They were conducted in full compliance with the bioethical requirements of Directive 2010/63/EU and Federal Law of 27.12.2018 N498-FZ (ed. of 27.12.2019) "On Responsible Treatment of Animals and Amendments to Certain Legislative Acts of the Russian Federation". Research was carried out with approval from the MSU bioethics committee (reference number № 157-d). The study was also carried out in compliance with ARRIVE guidelines. Experimental setup. For the study, birds were transferred to an experimental room (250 × 400 cm). Subjects were placed in a wire mesh cage (65 cm × 50 cm × 50 cm) with a wooden perch and a bowl of water. At the bottom of the front wall, there was a slit (50 cm × 9 cm) through which a wooden platform (35 cm × 20 cm) could be inserted into the cage (Fig. 1 ). A ceramic tray (10 cm × 10 cm) with a transparent plastic cup (3 cm high and 4 cm in diameter) glued to it for the bait (2 mealworm larvae) was placed between two wooden rails (20 cm × 1 cm) attached to the platform, to guide tray movement. Two metal loops were glued to the front edge of the ceramic tray 8 cm apart, through which a string was threaded. Specific characteristics of the strings are outlined in the descriptions of the particular experiments. The vast majority of the strings were striped, with alternating dark and light areas, in order to enhance the visibility of their movement within the loops. The platform with the tray was prepared out of the bird's sight (this was done below the level of the table on which the bird cage was situated). Prior to each trial, the platform was initially placed in front of the bird for 2 to 3 seconds, allowing the crow to see the string, but preventing the bird from reaching it. Then, the platform was partially slid into the cage, so the crow could reach the ends of the string. If the bird did not pull an end of a string within 2 minutes, the platform was removed from the cage. The intertrial interval was approximately 1 minute, which was the estimated time required to prepare the platform with the tray for the next trial. The number of trials per day varied considerably, depending on the bird's willingness to work. To eliminate the possibility of a "Clever Hans" error, an opaque plastic screen (60 cm × 45 cm) was placed between the experimenter and the crow, preventing visual contact between them. As the experimenter was unable to see the bird, they had to rely on the movement of the ceramic tray or the string to judge the outcome of each trial. The bird's choice was correct if the tray moved and incorrect if the string slipped. Familiarisation and pre-training. The crows were first habituated to being in the experimental cage and receiving a reward from the experimenter via the slit in the front wall. The birds were then provided with the opportunity to pull the baited tray in using a string (red with black stripes, 35 cm in length and 0.3 cm in diameter) that had been threaded through loops on the tray and then tied to the middle of the other end of the same string, which was placed on the platform between the slats (Fig. 2 a). The pre-training phase was concluded when each bird had pulled the tray 10 times. Experiment 1: the loose-string task with stoppers. In Experiment 1, we initially tested whether the birds could solve the task spontaneously. If they failed, we taught them to pull a tray in by a specific end of a loose string – the one to which no stopper was attached. The string (60 cm in length and 0.3 cm in diameter) was threaded through two loops on the tray, with both ends (25 cm in length) placed parallel between the slats. At one end of the string, 12 or 18 cm from it, an object (beads, buttons, etc.) was tied. The size of the object (the stopper) was larger than the diameter of the loop on the tray, so it kept the string from slipping out of the loop, but only if the bird pulled the end of the string without the stopper (Fig. 2 b). The ends of the string with the stopper were placed on the different sides in a semi-random manner. The tray could be retrieved by simultaneously pulling both ends of the string. For the analysis, only the initial choice was considered. The trial was considered successful if the bird initially selected the 'correct' end. If the bird initially selected the 'incorrect' end, the platform with the tray was removed from the bird's reach, and the trial was considered unsuccessful (Supplementary Video). In the first test session we tested whether the birds could spontaneously solve this task over the course of 30 trials. We used three types of strings (blue with dark blue stripes, red with black stripes, or pink with purple stripes), which were alternated in a semi-random manner, and 30 unique stoppers in each of the 30 trials (Fig. 2 e). If crows failed to complete the task, they were trained to pull the tray in (first training phase). A knot was used as stopper, tied at a distance of 12 cm from the end of the string (Fig. 2 f). The birds were trained until they were able to successfully retract the tray in 9 out of 10 consecutive trials (p = 0.011, one-tailed binomial test). To determine whether any generalisation of learning had occurred, subjects were given a second test session with 30 different stoppers (30 trials). In the event of a failure, the crow was trained (second training phase) with three new stoppers (Fig. 2 g) that were not among the 30 used previously. Training was conducted until the crows were able to successfully retrieve the tray in 20 out of 24 consecutive trials (p = 0.0008, one-tailed binomial test). Thereafter, the subjects were presented with the third test session with 30 different stoppers. Experiment 2: the standard loose-string task. In Experiment 2, we tested whether crows acquired an understanding of how a loose string works or if they just learned a simple associative rule (i.e., 'pull the end without the stopper') while completing the previous task. The standard loose-string task was used (Fig. 2 c). A light blue string with dark blue stripes (35 cm in length, with a diameter of 0.3 cm) was threaded through the loops on the tray. The tray could be retrieved only by simultaneously pulling on both ends of a string (Supplementary Video). We first tested whether the birds could complete the task spontaneously. If they could not, they were trained to pull the tray in. During training, the ends of the string were initially overlapped. Once the subject had pulled the tray in 9 times in a row, the distance between the strings was increased by 1 cm. The training was completed when the birds were able to pull the tray in by the two ends, with a distance of 5 cm between them. A further 30 trials were conducted to ascertain whether the birds had developed a motor pattern of selecting the left or right end first. Experiment 3: the loose-string task with an additional short string. In Experiment 3, we tested whether the experience gained during the preceding two tasks had led to an understanding of the principles underlying a loose-string task. In particular, we determined whether crows would select the ends of a string passed through loops in a tray if an additional short string not connected to the tray was placed parallel to them. In addition to a long light blue string with dark blue stripes (35 cm in length, with a diameter of 0.3 cm) threaded through loops on the tray, a short string (same colour and diameter, 11 cm in length) was placed on the platform (Fig. 2 d). The short string was placed in a semi-random manner on the sides or between the ends of the long string (parallel to them). The end of the short string was not attached to the ceramic tray and was positioned at a distance of 3 cm from its edge. Two test sessions, consisting of 30 trials each, were conducted. Video coding and analyses. All training and test trials were recorded using a video camera. Trials were filmed and coded in terms of subjects’ binary choices. Additionally, data was recorded manually at the time of training and testing. All trials in Experiment 3 were also coded according to the order in which the subject combined the strings. Data analysis was conducted using R software (version 4.0.3). Since only six subjects were tested, only within-subject data was statistically analysed. Individual performance within each condition was assessed using binomial tests. Specifically, the binom.test () function in R was used to determine if the observed success rates were significantly different from what would be expected by chance. The number of correct responses out of the total number of trials was tested against a chance level of 50% (Experiment 1) or 33% (Experiment 3; in this experiment, the tray could be pulled in by the two ends using two out of the six possible combinations of ends). The success rates achieved in the initial 30 trials of Experiment 1, along with the presence of the motor pattern of selecting the left or right end first in Experiment 2, were analysed using a two-tailed binomial test. We used a one-tailed binomial test to assess the success rates in the remaining trials of Experiment 1 and in Experiment 3, as we were testing a directional hypothesis that after the training crows would perform above chance. For comparing the performance of crows in two different sessions in Experiment 3, we applied the two-tailed paired Wilcoxon signed-rank test. The wilcox.test() function in R was used to calculate the test statistics and p-values, providing insights into the learning effect. This non-parametric test is appropriate for evaluating differences between two related samples and does not assume a normal distribution of the data. In experiment 3, to compare success rates and the frequency of use of a particular strategy between different types of trials, we used a two-tailed z-test based on the proportions of successes or frequencies. This method was particularly useful for cases with unequal trial numbers. The z-test was conducted by calculating the pooled proportion of successes and using this to determine the standard error. The z-score was computed using the formula: z=(p1​−p2​​)/SE, where p1​ and p2​ are the proportions of successes in each condition, and SE is the standard error of the difference between the two proportions. The significance of the z-score was evaluated to determine if the differences in success rates or frequencies were statistically significant. Results Familiarisation and pre-training. All birds retrieved the tray in each of the first 10 trials. Experiment 1: the loose-string task with stoppers. In the first test session, none of the six crows chose the 'correct' end (the one without a stopper) above chance. Glas did so in 19 out of 30 trials (p = 0.201, two-tailed binomial test), Rodya in 11(p = 0.201), Joe in 16 (p = 0.856), Clara and Schnobel in 17 trials (p = 0.585). Grisha chose the 'incorrect' end above chance (21/30, p = 0.043). Subsequently, the crows were trained to pull the tray in using the end of the string to which the stopper was not attached (first training phase). Initially, a knot was used as a stopper. Crows reached the criterion (retrieve the tray in 9 out of 10 consecutive trials) after 80 (Glaz), 309 (Grisha), 38 (Rodya), 70 (Joe), 18 (Clara), and 95 (Schnobel) trials. In the second test session three crows chose the 'incorrect' end above chance. Schnobel pulled the 'correct' end in 25 out of 30 trials (p = 0.0002, one-tailed binomial test), Glaz in 21 (p = 0.021), and Grisha in 20 (p = 0.049). The remaining three subjects were trained with three new stoppers (second training phase). Clara died. Rodya and Joe reached the criterion (retrieve the tray in 20 out of 24 consecutive trials) after 420 and 453 trials. In the third test session Joe selected the 'correct' end above chance (23/30, p = 0.003, one-tailed binomial test) and Rodya performed at chance (19/30, p = 0.10). However, he selected the 'correct' string above chance in the fourth test session, which was conducted immediately after the third (28/30, p < < 0.0001). Experiment 2: the standard loose-string task. In the initial 30 trials, none of the crows pulled the tray in. They did not attempt to bring the two ends together. Further training was conducted. Four crows reached the criterion (retrieve the tray in 9 out of 10 consecutive trials) after 150 (Glaz), 105 (Grisha), 152 (Rodya), and 243 (Joe) trials. Schnoebel failed to reach the criterion in 200 trials and subsequently died. The additional 30 trials conducted after the birds reached the criterion revealed that Grisha and Rodya displayed a distinct motor pattern of taking a specific (left or right) end first. In all 30 trials, Grisha took the left end of first, and Rodya took the right end first. The other two birds did not exhibit a distinct motor pattern: Glaz took the right or left end first in 16 and 14 trials, respectively (p = 0.856, two-tailed binomial test), while Joe took the right or left end first in 19 and 11 trials (p = 0.20). Experiment 3: the loose-string task with an additional short string. The results for each crow in each of the 60 trials are presented in the Supplementary Table S1 . In some trials, all four crows demonstrated an unexpected solution to the problem, pulling all three ends (Table 1 , Supplementary Video). Glaz used this strategy in 20% of trials, Grisha in 38.3%, and Rodya and Joe both in 43.3% of trials. Table 1 Experiment 3: the number of trials where crows pulled all three ends in different trial types. The data are presented for each of the three trial types and for all 60 trials. Trial type Number of trials in which crows pulled all three ends / total number of trials Glaz Grisha Rodya Joe Short string on the right 4/20 11/20 9/20 7 /20 Short string on the left 3/20 10/20 5/20 10/20 Short string in the middle 5/20 2/20 12/20 9/20 All three trial types 12/60 23/60 26/60 26/60 Grisha was significantly less likely to use this strategy in trials where the short string was in the middle compared to both trials where the short string was on the left (p = 0.006, two-tailed z-test) and on the right (p = 0.002). Rodya was significantly less likely to use this strategy in trials where the short string was on the left, compared to trials where the short string was in the middle (p = 0.025). With regard to the remaining birds, no clear effect of trial type on this behaviour was observed. Glaz and Joe used this strategy about equally often in both sessions (p = 0.072 and p = 0.346, paired two-tailed Wilcoxon Signed-Rank Test), and Rodya and Grisha did it more often in the first session (p = 0.006 and p = 0.002, Table 2 ). Table 2 Experiment 3: the number of trials in the first and second sessions where the crows pulled all three ends. Wilcoxon Signed-Rank Test (paired, two-tailed) results were used to compare the number of trials where crows pulled all three ends in the first and second sessions. Significant results are bold. Number of trials in which crows pulled all three ends / total number of trials Glaz Grisha Rodya Joe First session (1) 4/30 16/30 21/30 12/30 Second session (2) 8/30 7/30 5/30 14/30 Wilcoxon test results for (1) vs. (2) 0.072 0.006 0.002 0.346 In those trials where crows combined the two ends, only Rodya selected the 'correct' ends significantly above chance across both sessions (21/34, p = 0.0006, one-tailed binomial test, Table 3 ). Joe selected the 'correct' ends above chance in the first session, but not in the second. Table 3 Experiment 3: crow's performance in trials where birds pulled the two ends. The data is presented for both sessions and separately for each of the two sessions and for the first and last 10 trials. Individual performance was assessed using a binomial test (one-tailed) with chance level at 33%. Z-test (two-tailed) was used to compare the success rates in first and second sessions. Significant results are bold. Number of correct choices / number of trials Glaz Grisha Rodya Joe Both sessions 21/48 p = 0.086 13/37 p = 0.469 21/34 p = 0.0006 16/34 p = 0.067 First session (1) 13/26 p = 0.058 5/14 p = 0.525 7/9 p = 0.008 10/18 p = 0.043 Second session (2) 8/22 p = 0.456 8/23 p = 0.519 14/25 p = 0.016 6/16 p = 0.453 Z-test results for (1) vs. (2) 0.343 0.954 0.25 0.292 A comparison of the performance of the all crows in the first and second sessions revealed no significant effect of eхperimental learning on the performance of any of them (two-tailed z-test, Table 3 ). In contrast, Joe selected the 'correct' ends above chance in the first session (10 out of 18 trials, p = 0.043, one-tailed binomial test), but not in the second. The type of trial had a significant impact on the success of the task for all four crows (Table 4 ). During the trials in which a short string was in the middle, none of the crows selected the 'correct' ends of the long string above chance. Glaz did not pull the tray in any of the 15 trials. The performance of other birds was significantly worse in this type of trial compared to the other two types. Table 4 Experiment 3: crow's performance in trials in which birds pulled the two ends. The data are presented for each of the three trial types. Individual performance was assessed using a one-tailed binomial test with chance level at 33%. Z-test (two-tailed) was used to compare the success rates in first and second sessions. Significant results are bold. Trial type Number of correct choices / number of trials Glaz Grisha Rodya Joe Short string on right side (1) 13/16 p = 0.0001 6/9 p = 0.042 9/11 p = 0.001 7/13 p = 0.103 Short string on left side (2) 8/17 p = 0.172 5/10 p = 0.213 8/15 p = 0.088 7/10 p = 0.02 Short string in the middle (3) 0/15 — 2/18 p = 0.993 4/8 p = 0.259 2/11 p = 0.925 Z-test results for (1) vs. (2) 0.041 0.46 0.131 0.43 Z-test results for (1) vs. (3) 0.00005 0.003 0.14 0.072 Z-test results for (2) vs. (3) 0.002 0.023 0.88 0.017 In some trials, all four crows combined the left and right ends (Supplementary Table S1 , Supplementary Video). Glaz only did this once in the first trial, connecting the 'correct' end of the long string with the 'incorrect' end of the short string. Grisha connected the left and right ends in two trials, combining the 'correct' ends. Rodya did the same in five trials. Joe connected the left and right ends in seven trials, combining the 'correct' ends in three of them. Glaz, Grisha, and Rodya selected the 'correct' ends above chance when the short string was on the right, but not when it was on the left (Table 4 ). The z-test confirmed that Glaz's accuracy rates differed significantly (p = 0.041, two-tailed). Joe chose the 'correct' ends above chance only when the short string was on the left, but the z-test showed no difference in accuracy rates between the two types of trials. To assess the potential influence of a motor pattern on task performance, we analysed the number of trials in which birds took strings in the left-to-right direction and vice versa (Table 5 ). Grisha and Rodya displayed the same motor pattern in the third experiment as in the end of Experiment 2. Grisha took ends from left to right in 56 out of 60 trials and Rodya took ends from right to left in 59 out of 60 trials. Table 5 Experiment 3: the number of trials in which birds took strings in the left-to-right or right-to-left directions. The data are presented for all 60 trials. Direction of choice Number of trials Glaz Grisha Rodya Joe Left-to-right Left end taken first 29 40 1 22 Middle end taken first 11 16 0 13 Аltogether 40 56 1 35 Right-to-left Right end taken first 12 3 41 14 Middle end taken first 8 1 18 11 Аltogether 20 4 59 25 In all 60 trials, only Rodya selected one of the 'correct' ends first above chance (47/60, p = 0.034, one-tailed binomial test, Supplementary Table S1 ). However, he did so in the second session (26/30, p = 0.012), but not in the first (21/30, p = 0.432). The z-test indicated no significant difference in accuracy rates between the two sessions (p = 0.116, two-tailed z-test). Joe selected one of the 'correct' ends first above chance in the first session (25/30, p = 0.036), but not in the second (19/30). A more detailed analysis of Rodya's and Joe's behaviour across the three trial types was conducted to identify the factors that influenced their decision-making process (Table 6 ). Table 6 Experiment 3: Rodya's and Joe's performance. The data are presented for all 60 trials. In trials where crows pulled two ends, the number of successful trials out of the total number of trials is provided. Trial type Strategy used Rodya's performance Joy's performance Number of trials Number of trials Left first Middle first Right first Left first Middle first Right first Short string on right side Pulled two ends — 9/9 0/2 2/3 5/8 0/2 Pulled three ends — — 9 2 1 4 Аltogether — 9 11 5 9 6 Short string on left side Pulled two ends — 0/7 8/8 0/2 5/6 2/2 Pulled three ends — — 5 5 5 — Аltogether — 7 13 7 11 2 Short string in the middle Pulled two ends — 0/2 4/6 1/6 0/2 1/3 Pulled three ends 1 — 11 4 2 3 Аltogether 1 2 17 10 4 6 Total number of trials 1 18 41 22 24 14 The analysis further confirmed that Rodya's behaviour was strongly influenced by the motor pattern. In only one out of 60 trials did he take the left end first (the 'correct' end). He took the right end first in 41 trials, and the middle end first in 18 trials. He took the nearest left end second in 55 out of 60 trials. In trials in which Rodya took the middle end first, he always took the left end second. Despite the presence of a strong motor pattern, in trials where the 'incorrect' short string was on the right side, Rodya more often (p = 0.003, two-tailed z-test) took the 'correct' middle end first (9/11) when pulling the two ends. Rodya pulled all three ends more often (p = 0.0001, two-tailed z-test) in trials where one of the first ends taken was the wrong one. In only 6 out of the 26 trials in which he pulled all three ends, he initially combined the 'correct' ends and then took the end of the 'incorrect' short string (Supplementary Table S1 ). The analysis of Joe's performance revealed no clear influence of motor pattern on the choice of the first end (Table 6 ). He took the left and middle ends first in about the same number of trials, and only slightly less often started with the right end. Similarly to Rodya, he pulled all three ends more often in trials where one of the ends taken first was the wrong one (22 out of 26 trials, p < < 0.0001, two-tailed z-test). Discussion The objective of this study was to test whether Hooded crows could understand how a loose string works. Prior to the critical test (Experiment 3), the birds were presented with two different loose-string tasks (Experiment 1 and Experiment 2) to allow them to acquire some knowledge of the causal basis of the task. All three tasks were characterised by visible causality. The loops on the tray through which the string was passed were within sight of the birds. The strings had dark and light stripes, which may have facilitated tracking the movement of the string in the loops. The simple associative rules that the birds learnt when they were trained to solve one task were not applicable to the next. On the other hand, understanding the principle of the loose string in one task would allow for the solution of the subsequent task. This study demonstrates that all variants of the loose-string task, despite their apparent visible causality, proved challenging for the crows. None of the birds were able to spontaneously solve either the task with 30 different stoppers used in Experiment 1 or the standard loose-string task in Experiment 2. In Experiment 1, despite the positive perceptual-motor feedback from selecting the end of the string without a stopper, the 30 trials were insufficient for short-term learning. Despite hundreds of trials in Experiment 1, none of the crows found another way to solve the problem – they never attempted to bring the two ends together, i.e. to retrieve the tray in the second possible way (as in the standard loose-string task). In Experiment 1, three crows formed a sufficiently generalised rule while they were trained with a knot as a stopper, which allowed them to solve the task with 30 stoppers in the second test session. The other three birds were unable to generalise their learning to new strings with a new stopper. These birds required more than 400 trials to learn to solve the same task with three new stoppers and a more rigorous learning criterion. One of these crows (Joe) completed the test with 30 new stoppers in the third test session, and the second bird (Rodya) completed it in the fourth test session, without any additional training in between. One of the crows learnt to solve the task with the knot as a stopper in 18 trials, whereas the other required 309 trials. Some of the crows were able to complete the test session with 30 stoppers after training with a knot, while others had to learn how to solve the task with three different stoppers, which took them more than 400 trials. The results again revealed significant individual differences in learning ability between individuals, a finding that has been repeatedly noted in other studies of physical cognition (e.g. 44 , 45 , 46 , 47 ). The findings of Experiment 2 demonstrated that the birds had only previously acquired an associative rule to pull the end without the stopper, which was not applicable to solving the novel variant of the task. The birds required between 105 and 243 trials to learn to pull the tray in by two ends. A comparable number of trials was required for parrots and rooks to learn to solve the same task 11 , 12 , 17 , whereas chimpanzees required only 5 — 10 trials 10 . Two crows (Grisha and Rodya) exhibited a strong motor pattern of taking a particular end of the string first. A similar preference, known as side bias, is not uncommon in string pulling tasks and can affect performance of animals in a variety of tasks (e.g. 2 , 4 , 44 , 48 ). In Experiment 3, we tested whether the experience gained during the previous two tasks had resulted in the acquisition of some knowledge about the causal basis of the loose-string task. If the crows had acquired an understanding of how the loose string works, then they should be able to spontaneously solve the new task, in which a short string not connected to the tray was placed parallel to the ends of a long string passed through loops in a tray. In some trials (20% — 43% of all trials), all four crows demonstrated an unexpected solution to the problem: they pulled all three ends. This strategy was an effective approach because it did not require the selection of specific ends. However, in 60 trials, this strategy did not become the dominant approach in any of the birds. Indeed, two crows used this strategy more often in the first session (Table 2 ). In the trials where crows combined two ends, two crows (Rodya and Joe) selected the 'correct' ends above chance in the first session (Table 3 ). In both sessions, only Rodya did so. A comparison of his performance in the first and second sessions revealed no notable effect of learning on the test results. Joe's performance declined in the second session compared to the first, which may be attributed to fatigue-related errors and a potential loss of motivation. The success of problem solving in trials where crows combined two ends depended on the position of the short string. The performance of three birds (Glaz, Grisha and Joe) was significantly worse in trials where the short string was placed in the middle (Table 4 ). It seems probable that in Experiment 2, the birds may have formed a motor pattern of combining the neighbouring ends. The three crows (Glaz, Grisha and Rodya) demonstrated a preference for the 'correct' ends when the short string was positioned to the right but not when it was on the left. One bird (Joe) demonstrated an inverse result (Table 4 ). The motor pattern of taking first a particular end (left or right) has previously been identified in only two birds (Grisha and Rodya). The poor performance demonstrated by Rodya in trials in which the short string was positioned on the left did not fit his motor pattern of taking strings in the right-to-left direction (Table 5 , Table 6 ), according to which one would expect better performance in this type of the trials. Additional factors may have influenced the crows' performance in the task, including some knowledge of the causal basis of the task and reliance on perceptual-motor feedback. The latter may be either visual (the other end of a long string moves when the crow pulls the first end, motivating the bird to take it) or proprioceptive (the tension of the string that occurs when the crow picks up the two 'correct' ends). The importance of perceptual-motor feedback has been demonstrated in many studies, using a coiled string (e.g. 1 , 2 , 3 , 6 , 45 , 47 , 49 , 50 ), visually restricted conditions 51 , 52 , and weight-conditioned tasks 53 . A detailed examination of video recordings of Rodya's behaviour was conducted. We did not detect his explicit use of proprioceptive feedback. In all trials, he first brought the ends together without tensioning them and then abruptly pulled them, so that no ends moved prior to his actions and no tension arose (Supplementary Video). This differs from the behaviour of some other birds (e.g. Joe and Glaz), which could combine two ends, pull them slightly and then pick up the third end (Supplementary Video). Nevertheless, Rodya's reliance on the perceptual-motor feedback may have been subtle and unobvious when assessed through the recordings. The proprioceptive feedback did not serve as the sole determining factor, at least in the case of Rodya. To illustrate, in six out of the 26 trials in which he pulled all three ends, he initially combined two 'correct' ends and could perceive the string tension, yet still took the third 'incorrect' end. In the loose-string task perceptual feedback could only affect the choice of the second or third ends. Rodya's choice of the first end was strongly influenced by the motor pattern. In only one trial out of 60 did Rodya take the left end first. Despite the strong influence of the motor pattern, Rodya still selected one of the 'correct' ends first above chance (p = 0.034). However, he only did this in the second session (26/30, p = 0.012), not in the first (21/30, p = 0.432). This suggests he learned to avoid the end of the string not connected to the tray during the first session. Conversely, Joe chose one of the 'correct' ends first above chance only in the first session (p = 0.036). His result may indicate that some crows possessed an understanding of the non-funcularity of the string lacking connectivity with the tray. This assumption is consistent with our previous results, according to which some Hooded crows successfully solved the task with two baited strings, one of which was broken 44 . In conclusion, the experience acquired in Experiments 1 and Experiments 2 enabled two crows to successfully complete the loose-string task with the additional short string in Experiment 3. If the crows initially did not connect the two 'correct' ends, there was a greater probability that they would also take the third end. This suggests that the crows either relied on perceptual-motor feedback or had some knowledge of how a loose string works. The hypothesis that Rodya began to understand how a loose string works is supported by the observation that in the five trials, in which he connected the left and right ends, he did so exclusively with the 'correct' ends. Taken together, these findings suggest that some crows may have acquired some knowledge of the causal basis of the loose-string task. Although only two out of the four crows may have begun to understand how loose stringing works, this could indicate that this ability is present in this species of bird. It is possible that the crows' performance was also influenced by the perceptual feedback they received when they took the second or third end. Taylor and colleagues 2 proposed that a 'perceptual-motor feedback loop' and associative learning could underlie the successful solution of string pulling tasks. They also hypothesised that species with larger associative areas would integrate information between perceptual and motor pathways quicker than species with smaller associative brain areas. It was later shown that in animals with advanced cognitive abilities and high levels of brain complexity, a more complex interplay of cognitive factors may underlie the ability to solve string-pulling tasks 6 . It can be hypothesised that if the associative areas of the brain have developed sufficiently, perceptual feedback can initiate more complex cognitive processes, providing some knowledge of the causal basis of the experimental task 45 , 54 or, at least, for anticipating the consequences of their own actions, that is compatible with the 'embodied cognition' hypothesis 55 , 56 . In humans, it has been demonstrated that the beneficial effects of feedback stem from allowing people to adjust their strategies for performing the task and not from direct reinforcement mechanisms 57 . This study represents the first attempt to develop a set of tasks to assess the animals' ability to understand how the loose string works. The set of tasks can be further developed and improved. It may be beneficial to use coiled strings, which could help clarify the influence of perceptual feedback. Further studies are necessary to clarify the mechanisms underlying the solution of the loose-string tasks in animals of different species. The application of the set of string-pulling tasks provides animals with extensive experience with the paradigm, which may in turn lead to their understanding of the causal mechanisms underlying loose-string tasks. Cooperative tests with such animals may possibly provide a better understanding of the cognitive underpinnings of cooperative behaviour. Conversely, the cooperative tasks themselves can serve as an additional test for assessing understanding of how the loose string works. Declarations Author contributions This study was designed by AAS, KNK and MAC. Testing was conducted by KNK and MAC. AAS, KNK and MAC analysed and visualised the data. MAC coded the videos. The first draft of the manuscript was written by AAS and MAC and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript. Funding This study was supported by the Russian Science Foundation Project No. 23-28-00364. Compliance with ethical standards Conflict of interest All the authors declare that they have no conflict of interest. Ethical approval Research was carried out with approval from the MSU bioethics committee (reference number № 157-d). Data availability All supporting data is available as an electronic supplementary material. References Johnsson, R. D. et al. Wild Australian magpies learn to pull intact, not broken, strings to obtain food. Behav. Ecol. Sociobiol. 77 , 49 (2023). Taylor, A. H. et al . An investigation into the cognition behind spontaneous string pulling in new Caledonian crows. PLoS ONE 5 , e9345 (2010). Taylor, A. H., Knaebe, B. & Gray, R. D. An end to insight? New Caledonian crows can spontaneously solve problems without planning their actions. Proc. R. Soc. B. 279 , 4977–4981 (2012). Wang, L., Zhang, D. & Sui, J. Investigation of cognitive mechanisms and strategy on solving multiple string-pulling problems in Azure-winged magpie ( Cyanopica cyanus ). Anim. Cogn. 24 , 1–10 (2021). Wasserman, E. A., Nagasaka, Y., Castro, L. & Brzykcy, S. J. Pigeons learn virtual patterned-string problems in a computerized touch screen environment. Anim. Cogn. 16 , 737–753 (2013). Bastos, A. P. M., Wood, P. M. & Taylor, A. H. Kea ( Nestor notabilis ) fail a loose-string connectivity task. Sci. Rep. 11 , 15492 (2021). Jacobs, I. F. & Osvath, M. The string-pulling paradigm in comparative psychology. J. Comp. Psychol. 129 , 89–120 (2015). Bluff, L. A., Weir, A. A. S., Rutz, C., Wimpenny, J. H. & Kacelnik, A. Tool-related cognition in New Caledonian crows. Comp. Cogn. Behav. Rev. 2 , 1–25 (2007). Hare, B., Melis, A. P., Woods, V., Hastings, S. & Wrangham, R. Tolerance allows bonobos to outperform chimpanzees on a cooperative task. Curr. Biol. 17 , 619–623 (2007). Hirata, S. & Fuwa, K. Chimpanzees ( Pan troglodytes ) learn to act with other individuals in a cooperative task. Primates 48 , 13–21 (2007). Heaney, M., Gray, R. D. & Taylor, A. H. Keas perform similarly to chimpanzees and elephants when solving collaborative tasks. PLoS ONE 12 , e0169799 (2017). Seed, A. M., Clayton, N. S. & Emery, N. J. Cooperative problem solving in rooks ( Corvus frugilegus ). Proc. R. Soc. B. 275 , 1421–1429 (2008). Ostojić, L. & Clayton, N. S. Behavioural coordination of dogs in a cooperative problem-solving task with a conspecific and a human partner. Anim. Cogn. 17 , 445–459 (2014). Massen, J. J. M., Ritter, C. & Bugnyar, T. Tolerance and reward equity predict cooperation in ravens ( Corvus corax ). Sci. Rep. 5 , 15021 (2015). Asakawa-Haas, K., Schiestl, M., Bugnyar, T. & Massen, J. J. M. Partner choice in raven ( Corvus corax ) cooperation. PLoS ONE 11 , e0156962 (2016). Marshall-Pescini, S., Schwarz, J. F. L., Kostelnik, I., Virányi, Z. & Range, F. Importance of a species’ socioecology: Wolves outperform dogs in a conspecific cooperation task. Proc. Natl. Acad. Sci. USA 114 , 11793–11798 (2017). Tassin de Montaigu, C., Durdevic, K., Brucks, D., Krasheninnikova, A. & Von Bayern, A. Blue‐throated macaws ( Ara glaucogularis ) succeed in a cooperative task without coordinating their actions. Ethology 126 , 267–277 (2020). Schwing, R., Jocteur, E., Wein, A., Noë, R. & Massen, J. J. M. Kea cooperate better with sharing affiliates. Anim. Cogn. 19 , 1093–1102 (2016). Plotnik, J. M., Lair, R., Suphachoksahakun, W. & De Waal, F. B. M. Elephants know when they need a helping trunk in a cooperative task. Proc. Natl. Acad. Sci. USA 108 , 5116–5121 (2011). Balakhonov, D. & Rose, J. Crows rival monkeys in cognitive capacity. Sci. Rep. 7 , 1–8 (2017). Lazareva, O. F. et al. Transitive responding in hooded crows requires linearly ordered stimuli. J. Exp. Anal. Behav. 82 , 1–19 (2004). Magnotti, J. F., Katz, J. S., Wright, A. A. & Kelly, D. M. Superior abstract-concept learning by Clark’s nutcrackers ( Nucifraga columbiana ). Biol. Lett. 11 , 20150148 (2015). Magnotti, J. F., Wright, A. A., Leonard, K., Katz, J. S. & Kelly, D. M. Abstract-concept learning in Black-billed magpies ( Pica hudsonia ). Psychon. Bull. Rev. 24 , 431–435 (2017). Samuleeva, M. V. & Smirnova, A. A. Emergence of reflexivity relation without identity matching-to-sample training in hooded crows ( Corvus cornix ). Cogn. Neurosci. 65 , (2020). Smirnova, A. A., Lazareva, O. F. & Zorina, Z. A. Use of number by crows: Investigation by matching and oddity learning. J. Exp. Anal. Behav. 73 , 163–176 (2000). Smirnova, A. A., Zorina, Z. A., Obozova, T. A. & Wasserman, E. A. Crows spontaneously exhibit analogical reasoning. Curr. Biol. 25 , 256–260 (2015). Smirnova, A. A., Obozova, T. A., Zorina, Z. A. & Wasserman, E. A. How do crows and parrots come to spontaneously perceive relations-between-relations?. Curr. Opin. Behav. Sci. 37 , 109–117 (2021). Seed, A. M., Emery, N. J. & Clayton, N. S. Intelligence in corvids and apes: A case of convergent evolution?. Ethology 115 , 401–420 (2009). Wilson, B., Mackintosh, N. J. & Boakes, R. A. Transfer of relational rules in matching and oddity learning by pigeons and corvids. Q. J. Exp. Psychol. 37 , 313–332 (1985). Wright, A. A., Magnotti, J. F., Katz, J. S., Leonard, K. & Kelly, D. M. Concept learning set-size functions for Clark’s nutcrackers. J. Exp. Anal. Behav. 105 , 76–84 (2016). Wright, A. A. et al. Corvids outperform pigeons and primates in learning a basic concept. Psych. Sci. 28 , 437–444 (2017). Güntürkün, O. & Bugnyar, T. Cognition without Cortex. Trends Cogn. Sci. 20 , 291–303 (2016). Güntürkün, O., Ströckens, F., Scarf, D. & Colombo, M. Apes, feathered apes, and pigeons: differences and similarities. Curr. Opin. Behav. Sci. 16 , 35–40 (2017). Güntürkün, O., Von Eugen, K., Packheiser, J. & Pusch, R. Avian pallial circuits and cognition: A comparison to mammals. Curr. Opin. Neurobiol. 71 , 29–36 (2021). Sayol, F., Lefebvre, L. & Sol, D. Relative brain size and its relation with the associative pallium in birds. Brain Behav. Evol. 87 , 69–77 (2016). Ströckens, F. et al. High associative neuron numbers could drive cognitive performance in corvid species. J. Comp. Neurol. 530 , 1588–1605 (2022). Von Eugen, K., Tabrik, S., Güntürkün, O. & Ströckens, F. A comparative analysis of the dopaminergic innervation of the executive caudal nidopallium in pigeon, chicken, zebra finch, and carrion crow. J. Comp. Neurol. 528 , 2929–2955 (2020). Shumaker, R. W., Walkup, K. R. & Beck, B. B. Animal Tool Behavior: The Use and Manufacture of Tools by Animals (Johns Hopkins University Press, 2011). Davenport, J., O’Callaghan, M. J. A., Davenport, J. L. & Kelly, T. C. Mussel dropping by Carrion and Hooded crows: biomechanical and energetic considerations. J. Field Ornithol. 85 , 196–205 (2014). Heinrich, B. Raven Tool Use?. The Condor 90 , 270–271 (1988). Smirnova, A. A., Bulgakova, L. R., Cheplakova, M. A. & Jelbert, S. A. Hooded crows ( Corvus cornix ) manufacture objects relative to a mental template. Anim. Cogn. 27 , 36 (2024). Jelbert, S. A., Hosking, R. J., Taylor, A. H. & Gray, R. D. Mental template matching is a potential cultural transmission mechanism for New Caledonian crow tool manufacturing traditions. Sci. Rep. 8 , 8956 (2018). Laumer, I. B., Jelbert, S. A., Taylor, A. H., Rössler, T. & Auersperg, A. M. I. Object manufacture based on a memorized template: Goffin’s cockatoos attend to different model features. Anim. Cogn. 24 , 457–470 (2021). Bagotskaya, M. S., Smirnova, A. A. & Zorina, Z. A. Corvidae can understand logical structure in baited string-pulling tasks. Neurosci. Behav. Physi. 42 , 36–42 (2012). Hofmann, M. M., Cheke, L. G. & Clayton, N. S. Western scrub-jays ( Aphelocoma californica ) solve multiple-string problems by the spatial relation of string and reward. Anim. Cogn. 19 , 1103–1114 (2016). Riemer, S., Müller, C., Range, F. & Huber, L. Dogs ( Canis familiaris ) can learn to attend to connectivity in string pulling tasks. J. Comp. Psychol. 128 , 31–39 (2014). Wakonig, B., Auersperg, A. M. I. & O’Hara, M. String-pulling in the Goffin’s cockatoo ( Cacatua goffiniana ). Learn. Behav. 49 , 124–136 (2021). Baciadonna, L., Cornero, F. M., Clayton, N. S. & Emery, N. J. Mirror-mediated string-pulling task in Eurasian jays ( Garrulus glandarius ). Anim. Cogn. 25 , 691–700 (2022). Alem, S. et al. Associative mechanisms allow for social learning and cultural transmission of string pulling in an insect. PLoS Biol. 14 , e1002564 (2016). Danel, S., Von Bayern, A. M. P. & Osiurak, F. Ground-hornbills ( Bucorvus ) show means-end understanding in a horizontal two-string discrimination task. J. Ethol. 37 , 117–122 (2019). Gaycken, J., Picken, D. J., Pike, T. W., Burman, O. H. P. & Wilkinson, A. Mechanisms underlying string-pulling behaviour in green-winged macaws. Behaviour 156 , 619–631 (2019). Chaves Molina, A. B., Cullell, T. M. & Mimó, M. C. String-pulling in African grey parrots ( Psittacus erithacus ): performance in discrimination tasks. Behaviour 156 , 847–857 (2019). Schrauf, C. & Call, J. Great apes use weight as a cue to find hidden food. Am. J. Primatol. 73 , 323–334 (2011). Schmidt, G. F. & Cook, R. G. Mind the gap: Means–end discrimination by pigeons. Anim. Behav. 71 , 599–608 (2006). Wilson, M. Six views of embodied cognition. Psychon. Bull. Rev. 9 , 625–636 (2002). Gibbs, R. W. Embodiment and Cognitive Science (Cambridge University Press, 2006). Haddara, N. & Rahnev, D. The impact of feedback on perceptual decision-making and metacognition: Reduction in bias but no change in sensitivity. Psychol. Sci. 33 , 259–275 (2022). Additional Declarations No competing interests reported. Supplementary Files SupplementaryTableS1.xlsx SupplementaryVideo.mp4 Cite Share Download PDF Status: Published Journal Publication published 04 May, 2025 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Revision requested 05 Mar, 2025 Reviews received at journal 03 Mar, 2025 Reviewers agreed at journal 18 Feb, 2025 Reviews received at journal 17 Dec, 2024 Reviewers agreed at journal 26 Nov, 2024 Reviewers agreed at journal 23 Nov, 2024 Reviewers invited by journal 21 Nov, 2024 Editor assigned by journal 21 Nov, 2024 Editor invited by journal 21 Nov, 2024 Submission checks completed at journal 20 Nov, 2024 First submitted to journal 06 Nov, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-5405438","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":382893024,"identity":"7dd87474-2489-4e9e-a13c-f46398af238a","order_by":0,"name":"Anna A. Smirnova","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAyklEQVRIiWNgGAWjYNACAwYGNvYGEMOCBC18PAdADAkSLJKTSABRRGjhn917+NONgnt5bJLPr274USDBwN/enYBXi8Sdc2nSOQbFxWzSOWU3e4AOkzhzdgN+a27kmDHnGCQktknnpN3gAWoxkMjFr0X+Ro7xZ7AWyTNpN/8Qo8XgRo6BNFiLBPux20TZYgh0GEQLTw7bbRkDCR6CfpEDO+xPQuL89uPPbr75YyPH395LwPsIwGMAJolVDgLsD0hRPQpGwSgYBSMIAABvjkOuMSuOQAAAAABJRU5ErkJggg==","orcid":"","institution":"Lomonosov Moscow State University","correspondingAuthor":true,"prefix":"","firstName":"Anna","middleName":"A.","lastName":"Smirnova","suffix":""},{"id":382893025,"identity":"b0ec3367-8fd6-4232-a383-e5eee0cc7283","order_by":1,"name":"Maria A. Cheplakova","email":"","orcid":"","institution":"Lomonosov Moscow State University","correspondingAuthor":false,"prefix":"","firstName":"Maria","middleName":"A.","lastName":"Cheplakova","suffix":""},{"id":382893027,"identity":"cbee3de7-b2ca-48b2-8e69-f2192dd42eff","order_by":2,"name":"Ksenia N. Kubenko","email":"","orcid":"","institution":"Lomonosov Moscow State University","correspondingAuthor":false,"prefix":"","firstName":"Ksenia","middleName":"N.","lastName":"Kubenko","suffix":""}],"badges":[],"createdAt":"2024-11-06 21:38:25","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5405438/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5405438/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-025-99778-z","type":"published","date":"2025-05-04T15:56:59+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":71495782,"identity":"ce789210-8ee9-4bfd-9f12-21b64013410b","added_by":"auto","created_at":"2024-12-16 08:15:01","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":2300072,"visible":true,"origin":"","legend":"\u003cp\u003eThe view of the experimental setup in Experiment 2, where crow had to pull both ends of a string simultaneously to retrieve the tray.\u003c/p\u003e","description":"","filename":"Fig1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-5405438/v1/aba0a584e4d286b70d640de7.jpg"},{"id":71495781,"identity":"b785f61c-5072-40f8-a2ea-56813d8e2192","added_by":"auto","created_at":"2024-12-16 08:15:01","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1215357,"visible":true,"origin":"","legend":"\u003cp\u003eDiagram of the platforms with the tray and the strings with stoppers used in Experiment 1: (a) Pre-training, (b) Experiment 1: the loose-string task with stopper, (c) Experiment 2:\u003cstrong\u003e \u003c/strong\u003ethe standard loose-string task, (d) Experiment 3: the loose-string task with an additional short string; (e) 30 strings with unique stoppers for the test sessions, (f) one string with a knot used in the first training phase, (g) three strings with additional new stoppers used in the second training phase.\u003c/p\u003e","description":"","filename":"Fig2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-5405438/v1/2ae6586a8ad6fef4481b4c9b.jpg"},{"id":81987438,"identity":"bb8853c3-3b0b-444f-97f4-ed56392d4146","added_by":"auto","created_at":"2025-05-05 16:02:24","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4683116,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5405438/v1/b68c430f-010a-4af0-baf3-921c2fbcb59c.pdf"},{"id":71495783,"identity":"1742b068-af48-404b-b0ad-ed2c2f3bc6b8","added_by":"auto","created_at":"2024-12-16 08:15:01","extension":"xlsx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":69665,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryTableS1.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-5405438/v1/b161c443404474ccde9f4f92.xlsx"},{"id":71495785,"identity":"e6c2f074-9022-415c-9bc0-93e5fe474e3a","added_by":"auto","created_at":"2024-12-16 08:15:02","extension":"mp4","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":50361845,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryVideo.mp4","url":"https://assets-eu.researchsquare.com/files/rs-5405438/v1/f36836fbded8fc9d89545d3b.mp4"}],"financialInterests":"No competing interests reported.","formattedTitle":"Do Hooded crows (Corvus cornix) understand how a loose string works? ","fulltext":[{"header":"Introduction","content":"\u003cp\u003eIn the domain of physical cognition, one of the most widely used paradigms is the string-pulling task (e.g.\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e,\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e,\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e), in which the animal must pull the baited string in order to obtain the bait. Even when using a series of such tasks of varying complexity, it's challenging to distinguish between solutions produced by reinforcement of random behaviour (i.e. induced by associative learning) and those resulting from understanding or reasoning\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e. Bluff and colleagues\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e were correct to point out that it is challenging to distinguish between associatively learned behaviour and associatively learned knowledge.\u003c/p\u003e \u003cp\u003eA variation of the string-pulling task can be considered a loose-string task, in which an out-of-reach tray baited with food rewards can be retrieved only by simultaneously pulling on both ends of a string threaded through the loops on the tray. Thе task, in which each of the two animals only has access to one end of the string, is commonly used for studies on animal cooperation. Some studies use the loose-string task in a pre-training phase, during which animals are individually taught to pull both ends of the string simultaneously\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e,\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e,\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e,\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e,\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e,\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e,\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e. Usually, no additional tests are conducted to determine whether the animals have understood how the loose string works. We found a single study that conducted a transfer test at the end of the training phase to assess whether the animals could generalise from the training trials to complete a task that presented the string in a novel manner\u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eUnderstanding of the causal basis of the loose-string task has rarely but been mentioned (e.g.\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e) when discussing factors influencing the outcome of a cooperative task. Although there is another point of view that the ability of the animals to learn the role of their partners neither implies nor requires that they understood how the loose string and the apparatus itself worked (e.g.\u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e).\u003c/p\u003e \u003cp\u003eIt is conceivable that a lack of knowledge of the causal basis of the loose-string task could make it more challenging to understand how the partner can assist with it. In the current study, we investigated whether Hooded crows (\u003cem\u003eCorvus cornix\u003c/em\u003e) could acquire some knowledge of the causal basis of the loose-string task.\u003c/p\u003e \u003cp\u003eСrows possess advanced cognitive abilities. They can extract relations among items and between relations, form abstract categories not tied to specific perceptual features and use abstract representations\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e,\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e,\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e,\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e,\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e,\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e,\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e,\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e,\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e,\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e,\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e,\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThe advanced cognitive abilities of corvids are determined by the high level of their brain complexity\u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e,\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e,\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e. They are characterised by an expanded associative meso- and nidopallium\u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e,\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u003c/sup\u003e. Meso- and nidopallium of corvids (at least in Carrion crows) is more densely and diversely innervated by dopaminergic fibres and they have more expended nidopallium caudolaterale \u0026mdash; the functional analogue to the mammalian prefrontal cortex\u003csup\u003e\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eHooded crows are not specialised tool users, but like other members of the corvid family\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e,\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e, they drop shells on rocks or nuts on motorways, which could be considered an example of proto-tool use. They throw twigs at intruders during the nesting season, although this may be a manifestation of displacement behaviour, which is typical for them when they are frustrated\u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eAs previously demonstrated by Bagotskaya et al. (2012), some Hooded crows are capable of solving sophisticated variants of horizontal string-pulling tasks for which the proximity rule is not applicable. Hauever, it remains unclear whether crows understand the principles underlying a string-pulling task or whether this result is due to rapid associative learning by using perceptual-motor feedback\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e,\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eWe have recently found\u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e that Hooded crows similar to tool-specialised New Caledonian crows\u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e and non-tool-specialised Goffin's cockatoos\u003csup\u003e\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u003c/sup\u003e, can manufacture objects according to a mental template.\u003c/p\u003e \u003cp\u003eIn the present study, a set of three tasks was applied, comprising one standard task (Experiment 2) and two novel ones. Prior to the critical test (Experiment 3), the birds were presented with two different loose-string tasks (Experiment 1 and Experiment 2) to allow them to acquire some knowledge of the causal basis of the task. In the initial two experiments, the ability of the crows to cope with the task spontaneously was assessed, and subsequently, the birds were trained to solve it if they failed. The third task was designed to assess whether the experience gained during the previous two tasks had resulted in the acquisition of some knowledge about the causal basis of the loose-string task.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e \u003cb\u003eSubjects.\u003c/b\u003e Six adult Hooded crows (\u003cem\u003eCorvus cornix\u003c/em\u003e) were tested. All of them were rescued from the wild due to injuries and were housed in the outdoor group aviary (500 \u0026times; 250 \u0026times; 300 cm) on the territory of the Lomonosov Moscow State University Botanical Garden. The aviaries are equipped with perches (tree branches, wooden ladders, stumps), wooden shelters, toys, metal and ceramic feeders and plastic basins with water. The birds are kept on an ad libitum diet (rat and mouse carcasses, steamed crops and seeds with added vegetable oil and vitamins, eggs, seasonal fruits and vegetables, and fresh water). If the crows refused to participate in the experiment, then they received food without animal protein for 1 or 2 days.\u003c/p\u003e \u003cp\u003eGlaz and Schnobel were kept for over 15 years, and Rodya, Joe, Grisha, and Clara for over four. The sex of the birds is not known exactly. Based on sex-size dimorphism and behaviour, it can be assumed that Grisha and Clara are females, and other crows are males. Prior to this experiment, all birds, except Grisha, had participated in experiments based on Aesop's fable paradigm. Two of the subjects (Glaz and Schnobel) had previously participated in a series of highly varied identity matching-to-sample tasks and later spontaneously performed relational matching-to-sample tasks\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e. Glaz and Schnobel participated in a string-pulling experiment\u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e. None of the other subjects had any prior experience with strings.\u003c/p\u003e \u003cp\u003eThe experiments were conducted from 2021 to 2023. All experiments were appetitive, non-invasive and based exclusively on behavioural tests. They were conducted in full compliance with the bioethical requirements of Directive 2010/63/EU and Federal Law of 27.12.2018 N498-FZ (ed. of 27.12.2019) \"On Responsible Treatment of Animals and Amendments to Certain Legislative Acts of the Russian Federation\". Research was carried out with approval from the MSU bioethics committee (reference number № 157-d). The study was also carried out in compliance with ARRIVE guidelines.\u003c/p\u003e \u003cp\u003e \u003cb\u003eExperimental setup.\u003c/b\u003e For the study, birds were transferred to an experimental room (250 \u0026times; 400 cm). Subjects were placed in a wire mesh cage (65 cm \u0026times; 50 cm \u0026times; 50 cm) with a wooden perch and a bowl of water. At the bottom of the front wall, there was a slit (50 cm \u0026times; 9 cm) through which a wooden platform (35 cm \u0026times; 20 cm) could be inserted into the cage (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eA ceramic tray (10 cm \u0026times; 10 cm) with a transparent plastic cup (3 cm high and 4 cm in diameter) glued to it for the bait (2 mealworm larvae) was placed between two wooden rails (20 cm \u0026times; 1 cm) attached to the platform, to guide tray movement. Two metal loops were glued to the front edge of the ceramic tray 8 cm apart, through which a string was threaded. Specific characteristics of the strings are outlined in the descriptions of the particular experiments. The vast majority of the strings were striped, with alternating dark and light areas, in order to enhance the visibility of their movement within the loops.\u003c/p\u003e \u003cp\u003eThe platform with the tray was prepared out of the bird's sight (this was done below the level of the table on which the bird cage was situated). Prior to each trial, the platform was initially placed in front of the bird for 2 to 3 seconds, allowing the crow to see the string, but preventing the bird from reaching it. Then, the platform was partially slid into the cage, so the crow could reach the ends of the string. If the bird did not pull an end of a string within 2 minutes, the platform was removed from the cage. The intertrial interval was approximately 1 minute, which was the estimated time required to prepare the platform with the tray for the next trial. The number of trials per day varied considerably, depending on the bird's willingness to work.\u003c/p\u003e \u003cp\u003eTo eliminate the possibility of a \"Clever Hans\" error, an opaque plastic screen (60 cm \u0026times; 45 cm) was placed between the experimenter and the crow, preventing visual contact between them. As the experimenter was unable to see the bird, they had to rely on the movement of the ceramic tray or the string to judge the outcome of each trial. The bird's choice was correct if the tray moved and incorrect if the string slipped.\u003c/p\u003e \u003cp\u003e \u003cb\u003eFamiliarisation and pre-training.\u003c/b\u003e The crows were first habituated to being in the experimental cage and receiving a reward from the experimenter via the slit in the front wall. The birds were then provided with the opportunity to pull the baited tray in using a string (red with black stripes, 35 cm in length and 0.3 cm in diameter) that had been threaded through loops on the tray and then tied to the middle of the other end of the same string, which was placed on the platform between the slats (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea). The pre-training phase was concluded when each bird had pulled the tray 10 times.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eExperiment 1: the loose-string task with stoppers.\u003c/b\u003e In Experiment 1, we initially tested whether the birds could solve the task spontaneously. If they failed, we taught them to pull a tray in by a specific end of a loose string \u0026ndash; the one to which no stopper was attached.\u003c/p\u003e \u003cp\u003eThe string (60 cm in length and 0.3 cm in diameter) was threaded through two loops on the tray, with both ends (25 cm in length) placed parallel between the slats. At one end of the string, 12 or 18 cm from it, an object (beads, buttons, etc.) was tied. The size of the object (the stopper) was larger than the diameter of the loop on the tray, so it kept the string from slipping out of the loop, but only if the bird pulled the end of the string without the stopper (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb). The ends of the string with the stopper were placed on the different sides in a semi-random manner. The tray could be retrieved by simultaneously pulling both ends of the string.\u003c/p\u003e \u003cp\u003eFor the analysis, only the initial choice was considered. The trial was considered successful if the bird initially selected the 'correct' end. If the bird initially selected the 'incorrect' end, the platform with the tray was removed from the bird's reach, and the trial was considered unsuccessful (Supplementary Video).\u003c/p\u003e \u003cp\u003eIn the first test session we tested whether the birds could spontaneously solve this task over the course of 30 trials. We used three types of strings (blue with dark blue stripes, red with black stripes, or pink with purple stripes), which were alternated in a semi-random manner, and 30 unique stoppers in each of the 30 trials (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ee).\u003c/p\u003e \u003cp\u003eIf crows failed to complete the task, they were trained to pull the tray in (first training phase). A knot was used as stopper, tied at a distance of 12 cm from the end of the string (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ef). The birds were trained until they were able to successfully retract the tray in 9 out of 10 consecutive trials (p\u0026thinsp;=\u0026thinsp;0.011, one-tailed binomial test).\u003c/p\u003e \u003cp\u003eTo determine whether any generalisation of learning had occurred, subjects were given a second test session with 30 different stoppers (30 trials).\u003c/p\u003e \u003cp\u003eIn the event of a failure, the crow was trained (second training phase) with three new stoppers (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eg) that were not among the 30 used previously. Training was conducted until the crows were able to successfully retrieve the tray in 20 out of 24 consecutive trials (p\u0026thinsp;=\u0026thinsp;0.0008, one-tailed binomial test).\u003c/p\u003e \u003cp\u003eThereafter, the subjects were presented with the third test session with 30 different stoppers.\u003c/p\u003e \u003cp\u003e \u003cb\u003eExperiment 2: the standard loose-string task.\u003c/b\u003e In Experiment 2, we tested whether crows acquired an understanding of how a loose string works or if they just learned a simple associative rule (i.e., 'pull the end without the stopper') while completing the previous task.\u003c/p\u003e \u003cp\u003eThe standard loose-string task was used (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ec). A light blue string with dark blue stripes (35 cm in length, with a diameter of 0.3 cm) was threaded through the loops on the tray. The tray could be retrieved only by simultaneously pulling on both ends of a string (Supplementary Video).\u003c/p\u003e \u003cp\u003eWe first tested whether the birds could complete the task spontaneously. If they could not, they were trained to pull the tray in. During training, the ends of the string were initially overlapped. Once the subject had pulled the tray in 9 times in a row, the distance between the strings was increased by 1 cm. The training was completed when the birds were able to pull the tray in by the two ends, with a distance of 5 cm between them.\u003c/p\u003e \u003cp\u003eA further 30 trials were conducted to ascertain whether the birds had developed a motor pattern of selecting the left or right end first.\u003c/p\u003e \u003cp\u003e \u003cb\u003eExperiment 3: the loose-string task with an additional short string.\u003c/b\u003e In Experiment 3, we tested whether the experience gained during the preceding two tasks had led to an understanding of the principles underlying a loose-string task. In particular, we determined whether crows would select the ends of a string passed through loops in a tray if an additional short string not connected to the tray was placed parallel to them.\u003c/p\u003e \u003cp\u003eIn addition to a long light blue string with dark blue stripes (35 cm in length, with a diameter of 0.3 cm) threaded through loops on the tray, a short string (same colour and diameter, 11 cm in length) was placed on the platform (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ed). The short string was placed in a semi-random manner on the sides or between the ends of the long string (parallel to them). The end of the short string was not attached to the ceramic tray and was positioned at a distance of 3 cm from its edge. Two test sessions, consisting of 30 trials each, were conducted.\u003c/p\u003e \u003cp\u003e \u003cb\u003eVideo coding and analyses.\u003c/b\u003e All training and test trials were recorded using a video camera. Trials were filmed and coded in terms of subjects\u0026rsquo; binary choices. Additionally, data was recorded manually at the time of training and testing. All trials in Experiment 3 were also coded according to the order in which the subject combined the strings.\u003c/p\u003e \u003cp\u003eData analysis was conducted using R software (version 4.0.3). Since only six subjects were tested, only within-subject data was statistically analysed. Individual performance within each condition was assessed using binomial tests.\u003c/p\u003e \u003cp\u003eSpecifically, the binom.test () function in R was used to determine if the observed success rates were significantly different from what would be expected by chance. The number of correct responses out of the total number of trials was tested against a chance level of 50% (Experiment 1) or 33% (Experiment 3; in this experiment, the tray could be pulled in by the two ends using two out of the six possible combinations of ends).\u003c/p\u003e \u003cp\u003eThe success rates achieved in the initial 30 trials of Experiment 1, along with the presence of the motor pattern of selecting the left or right end first in Experiment 2, were analysed using a two-tailed binomial test.\u003c/p\u003e \u003cp\u003eWe used a one-tailed binomial test to assess the success rates in the remaining trials of Experiment 1 and in Experiment 3, as we were testing a directional hypothesis that after the training crows would perform above chance.\u003c/p\u003e \u003cp\u003eFor comparing the performance of crows in two different sessions in Experiment 3, we applied the two-tailed paired Wilcoxon signed-rank test. The wilcox.test() function in R was used to calculate the test statistics and p-values, providing insights into the learning effect. This non-parametric test is appropriate for evaluating differences between two related samples and does not assume a normal distribution of the data.\u003c/p\u003e \u003cp\u003eIn experiment 3, to compare success rates and the frequency of use of a particular strategy between different types of trials, we used a two-tailed z-test based on the proportions of successes or frequencies. This method was particularly useful for cases with unequal trial numbers. The z-test was conducted by calculating the pooled proportion of successes and using this to determine the standard error. The z-score was computed using the formula: z=(p1​\u0026minus;p2​​)/SE, where p1​ and p2​ are the proportions of successes in each condition, and SE is the standard error of the difference between the two proportions. The significance of the z-score was evaluated to determine if the differences in success rates or frequencies were statistically significant.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e \u003cb\u003eFamiliarisation and pre-training.\u003c/b\u003e All birds retrieved the tray in each of the first 10 trials.\u003c/p\u003e \u003cp\u003e \u003cb\u003eExperiment 1: the loose-string task with stoppers.\u003c/b\u003e In the first test session, none of the six crows chose the 'correct' end (the one without a stopper) above chance. Glas did so in 19 out of 30 trials (p\u0026thinsp;=\u0026thinsp;0.201, two-tailed binomial test), Rodya in 11(p\u0026thinsp;=\u0026thinsp;0.201), Joe in 16 (p\u0026thinsp;=\u0026thinsp;0.856), Clara and Schnobel in 17 trials (p\u0026thinsp;=\u0026thinsp;0.585). Grisha chose the 'incorrect' end above chance (21/30, p\u0026thinsp;=\u0026thinsp;0.043).\u003c/p\u003e \u003cp\u003eSubsequently, the crows were trained to pull the tray in using the end of the string to which the stopper was not attached (first training phase). Initially, a knot was used as a stopper. Crows reached the criterion (retrieve the tray in 9 out of 10 consecutive trials) after 80 (Glaz), 309 (Grisha), 38 (Rodya), 70 (Joe), 18 (Clara), and 95 (Schnobel) trials.\u003c/p\u003e \u003cp\u003eIn the second test session three crows chose the 'incorrect' end above chance. Schnobel pulled the 'correct' end in 25 out of 30 trials (p\u0026thinsp;=\u0026thinsp;0.0002, one-tailed binomial test), Glaz in 21 (p\u0026thinsp;=\u0026thinsp;0.021), and Grisha in 20 (p\u0026thinsp;=\u0026thinsp;0.049).\u003c/p\u003e \u003cp\u003eThe remaining three subjects were trained with three new stoppers (second training phase). Clara died. Rodya and Joe reached the criterion (retrieve the tray in 20 out of 24 consecutive trials) after 420 and 453 trials.\u003c/p\u003e \u003cp\u003eIn the third test session Joe selected the 'correct' end above chance (23/30, p\u0026thinsp;=\u0026thinsp;0.003, one-tailed binomial test) and Rodya performed at chance (19/30, p\u0026thinsp;=\u0026thinsp;0.10). However, he selected the 'correct' string above chance in the fourth test session, which was conducted immediately after the third (28/30, p\u0026thinsp;\u0026lt;\u0026thinsp;\u0026lt;\u0026thinsp;0.0001).\u003c/p\u003e \u003cp\u003e \u003cb\u003eExperiment 2: the standard loose-string task.\u003c/b\u003e In the initial 30 trials, none of the crows pulled the tray in. They did not attempt to bring the two ends together.\u003c/p\u003e \u003cp\u003eFurther training was conducted. Four crows reached the criterion (retrieve the tray in 9 out of 10 consecutive trials) after 150 (Glaz), 105 (Grisha), 152 (Rodya), and 243 (Joe) trials. Schnoebel failed to reach the criterion in 200 trials and subsequently died.\u003c/p\u003e \u003cp\u003eThe additional 30 trials conducted after the birds reached the criterion revealed that Grisha and Rodya displayed a distinct motor pattern of taking a specific (left or right) end first. In all 30 trials, Grisha took the left end of first, and Rodya took the right end first. The other two birds did not exhibit a distinct motor pattern: Glaz took the right or left end first in 16 and 14 trials, respectively (p\u0026thinsp;=\u0026thinsp;0.856, two-tailed binomial test), while Joe took the right or left end first in 19 and 11 trials (p\u0026thinsp;=\u0026thinsp;0.20).\u003c/p\u003e \u003cp\u003e \u003cb\u003eExperiment 3: the loose-string task with an additional short string.\u003c/b\u003e The results for each crow in each of the 60 trials are presented in the Supplementary Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e.\u003c/p\u003e \u003cp\u003eIn some trials, all four crows demonstrated an unexpected solution to the problem, pulling all three ends (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, Supplementary Video). Glaz used this strategy in 20% of trials, Grisha in 38.3%, and Rodya and Joe both in 43.3% of trials.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eExperiment 3: the number of trials where crows pulled all three ends in different trial types. The data are presented for each of the three trial types and for all 60 trials.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eTrial type\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"4\" nameend=\"c5\" namest=\"c2\"\u003e \u003cp\u003eNumber of trials in which crows pulled all three ends\u003c/p\u003e \u003cp\u003e/ total number of trials\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlaz\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eGrisha\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRodya\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eJoe\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eShort string on the right\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4/20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e11/20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e9/20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e7 /20\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eShort string on the left\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3/20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e10/20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e5/20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e10/20\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eShort string in the middle\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e5/20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2/20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e12/20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e9/20\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAll three trial types\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e12/60\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e23/60\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e26/60\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e26/60\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eGrisha was significantly less likely to use this strategy in trials where the short string was in the middle compared to both trials where the short string was on the left (p\u0026thinsp;=\u0026thinsp;0.006, two-tailed z-test) and on the right (p\u0026thinsp;=\u0026thinsp;0.002). Rodya was significantly less likely to use this strategy in trials where the short string was on the left, compared to trials where the short string was in the middle (p\u0026thinsp;=\u0026thinsp;0.025). With regard to the remaining birds, no clear effect of trial type on this behaviour was observed.\u003c/p\u003e \u003cp\u003eGlaz and Joe used this strategy about equally often in both sessions (p\u0026thinsp;=\u0026thinsp;0.072 and p\u0026thinsp;=\u0026thinsp;0.346, paired two-tailed Wilcoxon Signed-Rank Test), and Rodya and Grisha did it more often in the first session (p\u0026thinsp;=\u0026thinsp;0.006 and p\u0026thinsp;=\u0026thinsp;0.002, Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eExperiment 3: the number of trials in the first and second sessions where the crows pulled all three ends. Wilcoxon Signed-Rank Test (paired, two-tailed) results were used to compare the number of trials where crows pulled all three ends in the first and second sessions. Significant results are bold.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"4\" nameend=\"c5\" namest=\"c2\"\u003e \u003cp\u003eNumber of trials in which crows pulled all three ends\u003c/p\u003e \u003cp\u003e/ total number of trials\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlaz\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eGrisha\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRodya\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eJoe\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFirst session (1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4/30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e16/30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e21/30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e12/30\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSecond session (2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e8/30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e7/30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e5/30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e14/30\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWilcoxon test results for (1) vs. (2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.072\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e0.006\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e0.002\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.346\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eIn those trials where crows combined the two ends, only Rodya selected the 'correct' ends significantly above chance across both sessions (21/34, p\u0026thinsp;=\u0026thinsp;0.0006, one-tailed binomial test, Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Joe selected the 'correct' ends above chance in the first session, but not in the second.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eExperiment 3: crow's performance in trials where birds pulled the two ends. The data is presented for both sessions and separately for each of the two sessions and for the first and last 10 trials. Individual performance was assessed using a binomial test (one-tailed) with chance level at 33%. Z-test (two-tailed) was used to compare the success rates in first and second sessions. Significant results are bold.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"4\" nameend=\"c5\" namest=\"c2\"\u003e \u003cp\u003eNumber of correct choices / number of trials\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlaz\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eGrisha\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRodya\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eJoe\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBoth sessions\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e21/48\u003c/p\u003e \u003cp\u003ep\u0026thinsp;=\u0026thinsp;0.086\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e13/37\u003c/p\u003e \u003cp\u003ep\u0026thinsp;=\u0026thinsp;0.469\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e21/34\u003c/b\u003e\u003c/p\u003e \u003cp\u003e\u003cb\u003ep\u0026thinsp;=\u0026thinsp;0.0006\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e16/34\u003c/p\u003e \u003cp\u003ep\u0026thinsp;=\u0026thinsp;0.067\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFirst session (1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e13/26\u003c/p\u003e \u003cp\u003ep\u0026thinsp;=\u0026thinsp;0.058\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e5/14\u003c/p\u003e \u003cp\u003ep\u0026thinsp;=\u0026thinsp;0.525\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e7/9\u003c/b\u003e\u003c/p\u003e \u003cp\u003e\u003cb\u003ep\u0026thinsp;=\u0026thinsp;0.008\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e10/18\u003c/b\u003e\u003c/p\u003e \u003cp\u003e\u003cb\u003ep\u0026thinsp;=\u0026thinsp;0.043\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSecond session (2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e8/22\u003c/p\u003e \u003cp\u003ep\u0026thinsp;=\u0026thinsp;0.456\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e8/23\u003c/p\u003e \u003cp\u003ep\u0026thinsp;=\u0026thinsp;0.519\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e14/25\u003c/b\u003e\u003c/p\u003e \u003cp\u003e\u003cb\u003ep\u0026thinsp;=\u0026thinsp;0.016\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e6/16\u003c/p\u003e \u003cp\u003ep\u0026thinsp;=\u0026thinsp;0.453\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZ-test results for (1) vs. (2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.343\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.954\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.25\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.292\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eA comparison of the performance of the all crows in the first and second sessions revealed no significant effect of eхperimental learning on the performance of any of them (two-tailed z-test, Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). In contrast, Joe selected the 'correct' ends above chance in the first session (10 out of 18 trials, p\u0026thinsp;=\u0026thinsp;0.043, one-tailed binomial test), but not in the second.\u003c/p\u003e \u003cp\u003eThe type of trial had a significant impact on the success of the task for all four crows (Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eDuring the trials in which a short string was in the middle, none of the crows selected the 'correct' ends of the long string above chance. Glaz did not pull the tray in any of the 15 trials. The performance of other birds was significantly worse in this type of trial compared to the other two types.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eExperiment 3: crow's performance in trials in which birds pulled the two ends. The data are presented for each of the three trial types. Individual performance was assessed using a one-tailed binomial test with chance level at 33%. Z-test (two-tailed) was used to compare the success rates in first and second sessions. Significant results are bold.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e \u003cp\u003eTrial type\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"4\" nameend=\"c5\" namest=\"c2\"\u003e \u003cp\u003eNumber of correct choices / number of trials\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eGlaz\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eGrisha\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRodya\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eJoe\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eShort string on right side (1)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e13/16\u003c/b\u003e\u003c/p\u003e \u003cp\u003e\u003cb\u003ep\u0026thinsp;=\u0026thinsp;0.0001\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e6/9\u003c/b\u003e\u003c/p\u003e \u003cp\u003e\u003cb\u003ep\u0026thinsp;=\u0026thinsp;0.042\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cb\u003e9/11\u003c/b\u003e\u003c/p\u003e \u003cp\u003e\u003cb\u003ep\u0026thinsp;=\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e7/13\u003c/p\u003e \u003cp\u003ep\u0026thinsp;=\u0026thinsp;0.103\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eShort string on left side (2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e8/17\u003c/p\u003e \u003cp\u003ep\u0026thinsp;=\u0026thinsp;0.172\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e5/10\u003c/p\u003e \u003cp\u003ep\u0026thinsp;=\u0026thinsp;0.213\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e8/15\u003c/p\u003e \u003cp\u003ep\u0026thinsp;=\u0026thinsp;0.088\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e7/10\u003c/b\u003e\u003c/p\u003e \u003cp\u003e\u003cb\u003ep\u0026thinsp;=\u0026thinsp;0.02\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eShort string in the middle (3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0/15\u003c/p\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2/18\u003c/p\u003e \u003cp\u003ep\u0026thinsp;=\u0026thinsp;0.993\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4/8\u003c/p\u003e \u003cp\u003ep\u0026thinsp;=\u0026thinsp;0.259\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e2/11\u003c/p\u003e \u003cp\u003ep\u0026thinsp;=\u0026thinsp;0.925\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZ-test results for (1) vs. (2)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e0.041\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.46\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.131\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.43\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZ-test results for (1) vs. (3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e0.00005\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e0.003\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.14\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e0.072\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZ-test results for (2) vs. (3)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cb\u003e0.002\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e0.023\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.88\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e0.017\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eIn some trials, all four crows combined the left and right ends (Supplementary Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e, Supplementary Video). Glaz only did this once in the first trial, connecting the 'correct' end of the long string with the 'incorrect' end of the short string. Grisha connected the left and right ends in two trials, combining the 'correct' ends. Rodya did the same in five trials. Joe connected the left and right ends in seven trials, combining the 'correct' ends in three of them.\u003c/p\u003e \u003cp\u003eGlaz, Grisha, and Rodya selected the 'correct' ends above chance when the short string was on the right, but not when it was on the left (Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). The z-test confirmed that Glaz's accuracy rates differed significantly (p\u0026thinsp;=\u0026thinsp;0.041, two-tailed). Joe chose the 'correct' ends above chance only when the short string was on the left, but the z-test showed no difference in accuracy rates between the two types of trials.\u003c/p\u003e \u003cp\u003eTo assess the potential influence of a motor pattern on task performance, we analysed the number of trials in which birds took strings in the left-to-right direction and vice versa (Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). Grisha and Rodya displayed the same motor pattern in the third experiment as in the end of Experiment 2. Grisha took ends from left to right in 56 out of 60 trials and Rodya took ends from right to left in 59 out of 60 trials.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eExperiment 3: the number of trials in which birds took strings in the left-to-right or right-to-left directions. The data are presented for all 60 trials.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"6\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"2\" morerows=\"1\" nameend=\"c2\" namest=\"c1\" rowspan=\"2\"\u003e \u003cp\u003eDirection of choice\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colspan=\"4\" nameend=\"c6\" namest=\"c3\"\u003e \u003cp\u003eNumber of trials\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eGlaz\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eGrisha\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eRodya\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eJoe\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003eLeft-to-right\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLeft end taken first\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e29\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e40\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e22\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMiddle end taken first\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e16\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e13\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eАltogether\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e40\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e56\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e35\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003eRight-to-left\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eRight end taken first\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e41\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e14\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMiddle end taken first\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e18\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e11\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eАltogether\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e20\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e59\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e25\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eIn all 60 trials, only Rodya selected one of the 'correct' ends first above chance (47/60, p\u0026thinsp;=\u0026thinsp;0.034, one-tailed binomial test, Supplementary Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). However, he did so in the second session (26/30, p\u0026thinsp;=\u0026thinsp;0.012), but not in the first (21/30, p\u0026thinsp;=\u0026thinsp;0.432). The z-test indicated no significant difference in accuracy rates between the two sessions (p\u0026thinsp;=\u0026thinsp;0.116, two-tailed z-test). Joe selected one of the 'correct' ends first above chance in the first session (25/30, p\u0026thinsp;=\u0026thinsp;0.036), but not in the second (19/30).\u003c/p\u003e \u003cp\u003eA more detailed analysis of Rodya's and Joe's behaviour across the three trial types was conducted to identify the factors that influenced their decision-making process (Table\u0026nbsp;\u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e6\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab6\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 6\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eExperiment 3: Rodya's and Joe's performance. The data are presented for all 60 trials. In trials where crows pulled two ends, the number of successful trials out of the total number of trials is provided.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"8\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003eTrial type\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003eStrategy used\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c5\" namest=\"c3\"\u003e \u003cp\u003eRodya's performance\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c8\" namest=\"c6\"\u003e \u003cp\u003eJoy's performance\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c5\" namest=\"c3\"\u003e \u003cp\u003eNumber of trials\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"3\" nameend=\"c8\" namest=\"c6\"\u003e \u003cp\u003eNumber of trials\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003cp\u003efirst\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMiddle\u003c/p\u003e \u003cp\u003efirst\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eRight\u003c/p\u003e \u003cp\u003efirst\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eLeft\u003c/p\u003e \u003cp\u003efirst\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c7\"\u003e \u003cp\u003eMiddle\u003c/p\u003e \u003cp\u003efirst\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c8\"\u003e \u003cp\u003eRight\u003c/p\u003e \u003cp\u003efirst\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003eShort string\u003c/p\u003e \u003cp\u003eon right side\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePulled two ends\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e9/9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0/2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e2/3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e5/8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e0/2\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePulled three ends\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eАltogether\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003eShort string\u003c/p\u003e \u003cp\u003eon left side\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePulled two ends\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0/7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e8/8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0/2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e5/6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e2/2\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePulled three ends\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eАltogether\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e \u003cp\u003eShort string\u003c/p\u003e \u003cp\u003ein the middle\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePulled two ends\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0/2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e4/6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1/6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0/2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e1/3\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePulled three ends\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u0026mdash;\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e3\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eАltogether\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e17\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e10\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e6\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTotal number of trials\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e18\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e41\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e22\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e24\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c8\"\u003e \u003cp\u003e14\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eThe analysis further confirmed that Rodya's behaviour was strongly influenced by the motor pattern. In only one out of 60 trials did he take the left end first (the 'correct' end). He took the right end first in 41 trials, and the middle end first in 18 trials. He took the nearest left end second in 55 out of 60 trials. In trials in which Rodya took the middle end first, he always took the left end second.\u003c/p\u003e \u003cp\u003eDespite the presence of a strong motor pattern, in trials where the 'incorrect' short string was on the right side, Rodya more often (p\u0026thinsp;=\u0026thinsp;0.003, two-tailed z-test) took the 'correct' middle end first (9/11) when pulling the two ends.\u003c/p\u003e \u003cp\u003eRodya pulled all three ends more often (p\u0026thinsp;=\u0026thinsp;0.0001, two-tailed z-test) in trials where one of the first ends taken was the wrong one. In only 6 out of the 26 trials in which he pulled all three ends, he initially combined the 'correct' ends and then took the end of the 'incorrect' short string (Supplementary Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe analysis of Joe's performance revealed no clear influence of motor pattern on the choice of the first end (Table\u0026nbsp;\u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). He took the left and middle ends first in about the same number of trials, and only slightly less often started with the right end. Similarly to Rodya, he pulled all three ends more often in trials where one of the ends taken first was the wrong one (22 out of 26 trials, p\u0026thinsp;\u0026lt;\u0026thinsp;\u0026lt;\u0026thinsp;0.0001, two-tailed z-test).\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe objective of this study was to test whether Hooded crows could understand how a loose string works. Prior to the critical test (Experiment 3), the birds were presented with two different loose-string tasks (Experiment 1 and Experiment 2) to allow them to acquire some knowledge of the causal basis of the task.\u003c/p\u003e \u003cp\u003eAll three tasks were characterised by visible causality. The loops on the tray through which the string was passed were within sight of the birds. The strings had dark and light stripes, which may have facilitated tracking the movement of the string in the loops.\u003c/p\u003e \u003cp\u003eThe simple associative rules that the birds learnt when they were trained to solve one task were not applicable to the next. On the other hand, understanding the principle of the loose string in one task would allow for the solution of the subsequent task.\u003c/p\u003e \u003cp\u003eThis study demonstrates that all variants of the loose-string task, despite their apparent visible causality, proved challenging for the crows. None of the birds were able to spontaneously solve either the task with 30 different stoppers used in Experiment 1 or the standard loose-string task in Experiment 2. In Experiment 1, despite the positive perceptual-motor feedback from selecting the end of the string without a stopper, the 30 trials were insufficient for short-term learning. Despite hundreds of trials in Experiment 1, none of the crows found another way to solve the problem \u0026ndash; they never attempted to bring the two ends together, i.e. to retrieve the tray in the second possible way (as in the standard loose-string task).\u003c/p\u003e \u003cp\u003eIn Experiment 1, three crows formed a sufficiently generalised rule while they were trained with a knot as a stopper, which allowed them to solve the task with 30 stoppers in the second test session. The other three birds were unable to generalise their learning to new strings with a new stopper. These birds required more than 400 trials to learn to solve the same task with three new stoppers and a more rigorous learning criterion. One of these crows (Joe) completed the test with 30 new stoppers in the third test session, and the second bird (Rodya) completed it in the fourth test session, without any additional training in between.\u003c/p\u003e \u003cp\u003eOne of the crows learnt to solve the task with the knot as a stopper in 18 trials, whereas the other required 309 trials. Some of the crows were able to complete the test session with 30 stoppers after training with a knot, while others had to learn how to solve the task with three different stoppers, which took them more than 400 trials. The results again revealed significant individual differences in learning ability between individuals, a finding that has been repeatedly noted in other studies of physical cognition (e.g.\u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e,\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e,\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e,\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e\u003c/sup\u003e).\u003c/p\u003e \u003cp\u003eThe findings of Experiment 2 demonstrated that the birds had only previously acquired an associative rule to pull the end without the stopper, which was not applicable to solving the novel variant of the task. The birds required between 105 and 243 trials to learn to pull the tray in by two ends. A comparable number of trials was required for parrots and rooks to learn to solve the same task\u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e, whereas chimpanzees required only 5 \u0026mdash; 10 trials\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eTwo crows (Grisha and Rodya) exhibited a strong motor pattern of taking a particular end of the string first. A similar preference, known as side bias, is not uncommon in string pulling tasks and can affect performance of animals in a variety of tasks (e.g.\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e,\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e,\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e\u003c/sup\u003e).\u003c/p\u003e \u003cp\u003eIn Experiment 3, we tested whether the experience gained during the previous two tasks had resulted in the acquisition of some knowledge about the causal basis of the loose-string task. If the crows had acquired an understanding of how the loose string works, then they should be able to spontaneously solve the new task, in which a short string not connected to the tray was placed parallel to the ends of a long string passed through loops in a tray.\u003c/p\u003e \u003cp\u003eIn some trials (20% \u0026mdash; 43% of all trials), all four crows demonstrated an unexpected solution to the problem: they pulled all three ends. This strategy was an effective approach because it did not require the selection of specific ends. However, in 60 trials, this strategy did not become the dominant approach in any of the birds. Indeed, two crows used this strategy more often in the first session (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn the trials where crows combined two ends, two crows (Rodya and Joe) selected the 'correct' ends above chance in the first session (Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). In both sessions, only Rodya did so. A comparison of his performance in the first and second sessions revealed no notable effect of learning on the test results. Joe's performance declined in the second session compared to the first, which may be attributed to fatigue-related errors and a potential loss of motivation.\u003c/p\u003e \u003cp\u003eThe success of problem solving in trials where crows combined two ends depended on the position of the short string. The performance of three birds (Glaz, Grisha and Joe) was significantly worse in trials where the short string was placed in the middle (Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). It seems probable that in Experiment 2, the birds may have formed a motor pattern of combining the neighbouring ends.\u003c/p\u003e \u003cp\u003eThe three crows (Glaz, Grisha and Rodya) demonstrated a preference for the 'correct' ends when the short string was positioned to the right but not when it was on the left. One bird (Joe) demonstrated an inverse result (Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). The motor pattern of taking first a particular end (left or right) has previously been identified in only two birds (Grisha and Rodya).\u003c/p\u003e \u003cp\u003eThe poor performance demonstrated by Rodya in trials in which the short string was positioned on the left did not fit his motor pattern of taking strings in the right-to-left direction (Table\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e, Table\u0026nbsp;\u003cspan refid=\"Tab6\" class=\"InternalRef\"\u003e6\u003c/span\u003e), according to which one would expect better performance in this type of the trials.\u003c/p\u003e \u003cp\u003eAdditional factors may have influenced the crows' performance in the task, including some knowledge of the causal basis of the task and reliance on perceptual-motor feedback. The latter may be either visual (the other end of a long string moves when the crow pulls the first end, motivating the bird to take it) or proprioceptive (the tension of the string that occurs when the crow picks up the two 'correct' ends). The importance of perceptual-motor feedback has been demonstrated in many studies, using a coiled string (e.g.\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e,\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e,\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e,\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e,\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e,\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u003c/sup\u003e), visually restricted conditions\u003csup\u003e\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e,\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e\u003c/sup\u003e, and weight-conditioned tasks\u003csup\u003e\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eA detailed examination of video recordings of Rodya's behaviour was conducted. We did not detect his explicit use of proprioceptive feedback. In all trials, he first brought the ends together without tensioning them and then abruptly pulled them, so that no ends moved prior to his actions and no tension arose (Supplementary Video). This differs from the behaviour of some other birds (e.g. Joe and Glaz), which could combine two ends, pull them slightly and then pick up the third end (Supplementary Video). Nevertheless, Rodya's reliance on the perceptual-motor feedback may have been subtle and unobvious when assessed through the recordings.\u003c/p\u003e \u003cp\u003eThe proprioceptive feedback did not serve as the sole determining factor, at least in the case of Rodya. To illustrate, in six out of the 26 trials in which he pulled all three ends, he initially combined two 'correct' ends and could perceive the string tension, yet still took the third 'incorrect' end.\u003c/p\u003e \u003cp\u003eIn the loose-string task perceptual feedback could only affect the choice of the second or third ends. Rodya's choice of the first end was strongly influenced by the motor pattern. In only one trial out of 60 did Rodya take the left end first. Despite the strong influence of the motor pattern, Rodya still selected one of the 'correct' ends first above chance (p\u0026thinsp;=\u0026thinsp;0.034). However, he only did this in the second session (26/30, p\u0026thinsp;=\u0026thinsp;0.012), not in the first (21/30, p\u0026thinsp;=\u0026thinsp;0.432). This suggests he learned to avoid the end of the string not connected to the tray during the first session. Conversely, Joe chose one of the 'correct' ends first above chance only in the first session (p\u0026thinsp;=\u0026thinsp;0.036). His result may indicate that some crows possessed an understanding of the non-funcularity of the string lacking connectivity with the tray. This assumption is consistent with our previous results, according to which some Hooded crows successfully solved the task with two baited strings, one of which was broken\u003csup\u003e\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eIn conclusion, the experience acquired in Experiments 1 and Experiments 2 enabled two crows to successfully complete the loose-string task with the additional short string in Experiment 3.\u003c/p\u003e \u003cp\u003eIf the crows initially did not connect the two 'correct' ends, there was a greater probability that they would also take the third end. This suggests that the crows either relied on perceptual-motor feedback or had some knowledge of how a loose string works. The hypothesis that Rodya began to understand how a loose string works is supported by the observation that in the five trials, in which he connected the left and right ends, he did so exclusively with the 'correct' ends. Taken together, these findings suggest that some crows may have acquired some knowledge of the causal basis of the loose-string task. Although only two out of the four crows may have begun to understand how loose stringing works, this could indicate that this ability is present in this species of bird.\u003c/p\u003e \u003cp\u003eIt is possible that the crows' performance was also influenced by the perceptual feedback they received when they took the second or third end. Taylor and colleagues\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e proposed that a 'perceptual-motor feedback loop' and associative learning could underlie the successful solution of string pulling tasks. They also hypothesised that species with larger associative areas would integrate information between perceptual and motor pathways quicker than species with smaller associative brain areas. It was later shown that in animals with advanced cognitive abilities and high levels of brain complexity, a more complex interplay of cognitive factors may underlie the ability to solve string-pulling tasks\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e. It can be hypothesised that if the associative areas of the brain have developed sufficiently, perceptual feedback can initiate more complex cognitive processes, providing some knowledge of the causal basis of the experimental task\u003csup\u003e\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e,\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e\u003c/sup\u003e or, at least, for anticipating the consequences of their own actions, that is compatible with the 'embodied cognition' hypothesis\u003csup\u003e\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e,\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e\u003c/sup\u003e. In humans, it has been demonstrated that the beneficial effects of feedback stem from allowing people to adjust their strategies for performing the task and not from direct reinforcement mechanisms\u003csup\u003e\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThis study represents the first attempt to develop a set of tasks to assess the animals' ability to understand how the loose string works. The set of tasks can be further developed and improved. It may be beneficial to use coiled strings, which could help clarify the influence of perceptual feedback. Further studies are necessary to clarify the mechanisms underlying the solution of the loose-string tasks in animals of different species.\u003c/p\u003e \u003cp\u003eThe application of the set of string-pulling tasks provides animals with extensive experience with the paradigm, which may in turn lead to their understanding of the causal mechanisms underlying loose-string tasks. Cooperative tests with such animals may possibly provide a better understanding of the cognitive underpinnings of cooperative behaviour. Conversely, the cooperative tasks themselves can serve as an additional test for assessing understanding of how the loose string works.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was designed by AAS, KNK and MAC. Testing was conducted by KNK and MAC. AAS, KNK and MAC analysed and visualised the data. MAC coded the videos. The first draft of the manuscript was written by AAS and MAC and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study was supported by the Russian Science Foundation Project No. 23-28-00364.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompliance with ethical standards\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll the authors declare that they have no conflict of interest.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthical approval\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eResearch was carried out with approval from the MSU bioethics committee (reference number № 157-d).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll supporting data is available as an electronic supplementary material.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eJohnsson, R. D. \u003cem\u003eet al.\u003c/em\u003e Wild Australian magpies learn to pull intact, not broken, strings to obtain food. \u003cem\u003eBehav. Ecol. Sociobiol. \u003c/em\u003e\u003cstrong\u003e77\u003c/strong\u003e, 49 (2023).\u003c/li\u003e\n\u003cli\u003eTaylor, A. H. \u003cem\u003eet al\u003c/em\u003e. An investigation into the cognition behind spontaneous string pulling in new Caledonian crows. \u003cem\u003ePLoS ONE\u003c/em\u003e \u003cstrong\u003e5\u003c/strong\u003e, e9345 (2010).\u003c/li\u003e\n\u003cli\u003eTaylor, A. H., Knaebe, B. \u0026amp; Gray, R. D. An end to insight? New Caledonian crows can spontaneously solve problems without planning their actions. \u003cem\u003eProc. R. Soc. B. \u003c/em\u003e\u003cstrong\u003e279\u003c/strong\u003e, 4977\u0026ndash;4981 (2012).\u003c/li\u003e\n\u003cli\u003eWang, L., Zhang, D. \u0026amp; Sui, J. Investigation of cognitive mechanisms and strategy on solving multiple string-pulling problems in Azure-winged magpie (\u003cem\u003eCyanopica cyanus\u003c/em\u003e). \u003cem\u003eAnim. Cogn. \u003c/em\u003e\u003cstrong\u003e24\u003c/strong\u003e, 1\u0026ndash;10 (2021).\u003c/li\u003e\n\u003cli\u003eWasserman, E. A., Nagasaka, Y., Castro, L. \u0026amp; Brzykcy, S. J. Pigeons learn virtual patterned-string problems in a computerized touch screen environment. \u003cem\u003eAnim. Cogn. \u003c/em\u003e\u003cstrong\u003e16\u003c/strong\u003e, 737\u0026ndash;753 (2013).\u003c/li\u003e\n\u003cli\u003eBastos, A. P. M., Wood, P. M. \u0026amp; Taylor, A. H. Kea (\u003cem\u003eNestor notabilis\u003c/em\u003e) fail a loose-string connectivity task. \u003cem\u003eSci. Rep. \u003c/em\u003e\u003cstrong\u003e11\u003c/strong\u003e, 15492 (2021).\u003c/li\u003e\n\u003cli\u003eJacobs, I. F. \u0026amp; Osvath, M. The string-pulling paradigm in comparative psychology. \u003cem\u003eJ. Comp. Psychol.\u003c/em\u003e \u003cstrong\u003e129\u003c/strong\u003e, 89\u0026ndash;120 (2015).\u003c/li\u003e\n\u003cli\u003eBluff, L. A., Weir, A. A. S., Rutz, C., Wimpenny, J. H. \u0026amp; Kacelnik, A. Tool-related cognition in New Caledonian crows. \u003cem\u003eComp. Cogn. Behav. Rev.\u003c/em\u003e \u003cstrong\u003e2\u003c/strong\u003e, 1\u0026ndash;25 (2007).\u003c/li\u003e\n\u003cli\u003eHare, B., Melis, A. P., Woods, V., Hastings, S. \u0026amp; Wrangham, R. Tolerance allows bonobos to outperform chimpanzees on a cooperative task. \u003cem\u003eCurr. Biol. \u003c/em\u003e\u003cstrong\u003e17\u003c/strong\u003e, 619\u0026ndash;623 (2007).\u003c/li\u003e\n\u003cli\u003eHirata, S. \u0026amp; Fuwa, K. Chimpanzees (\u003cem\u003ePan troglodytes\u003c/em\u003e) learn to act with other individuals in a cooperative task. \u003cem\u003ePrimates \u003c/em\u003e\u003cstrong\u003e48\u003c/strong\u003e, 13\u0026ndash;21 (2007).\u003c/li\u003e\n\u003cli\u003eHeaney, M., Gray, R. D. \u0026amp; Taylor, A. H. Keas perform similarly to chimpanzees and elephants when solving collaborative tasks. \u003cem\u003ePLoS ONE \u003c/em\u003e\u003cstrong\u003e12\u003c/strong\u003e, e0169799 (2017).\u003c/li\u003e\n\u003cli\u003eSeed, A. M., Clayton, N. S. \u0026amp; Emery, N. J. Cooperative problem solving in rooks (\u003cem\u003eCorvus frugilegus\u003c/em\u003e). \u003cem\u003eProc. R. Soc. B. \u003c/em\u003e\u003cstrong\u003e275\u003c/strong\u003e, 1421\u0026ndash;1429 (2008).\u003c/li\u003e\n\u003cli\u003eOstojić, L. \u0026amp; Clayton, N. S. Behavioural coordination of dogs in a cooperative problem-solving task with a conspecific and a human partner. \u003cem\u003eAnim. Cogn. \u003c/em\u003e\u003cstrong\u003e17\u003c/strong\u003e, 445\u0026ndash;459 (2014).\u003c/li\u003e\n\u003cli\u003eMassen, J. J. M., Ritter, C. \u0026amp; Bugnyar, T. Tolerance and reward equity predict cooperation in ravens (\u003cem\u003eCorvus corax\u003c/em\u003e). \u003cem\u003eSci. Rep. \u003c/em\u003e\u003cstrong\u003e5\u003c/strong\u003e, 15021 (2015).\u003c/li\u003e\n\u003cli\u003eAsakawa-Haas, K., Schiestl, M., Bugnyar, T. \u0026amp; Massen, J. J. M. Partner choice in raven (\u003cem\u003eCorvus corax\u003c/em\u003e) cooperation. \u003cem\u003ePLoS ONE \u003c/em\u003e\u003cstrong\u003e11\u003c/strong\u003e, e0156962 (2016).\u003c/li\u003e\n\u003cli\u003eMarshall-Pescini, S., Schwarz, J. F. L., Kostelnik, I., Vir\u0026aacute;nyi, Z. \u0026amp; Range, F. Importance of a species\u0026rsquo; socioecology: Wolves outperform dogs in a conspecific cooperation task. \u003cem\u003eProc. Natl. Acad. Sci. USA\u003c/em\u003e \u003cstrong\u003e114\u003c/strong\u003e, 11793\u0026ndash;11798 (2017).\u003c/li\u003e\n\u003cli\u003eTassin de Montaigu, C., Durdevic, K., Brucks, D., Krasheninnikova, A. \u0026amp; Von Bayern, A. Blue‐throated macaws (\u003cem\u003eAra glaucogularis\u003c/em\u003e) succeed in a cooperative task without coordinating their actions. \u003cem\u003eEthology \u003c/em\u003e\u003cstrong\u003e126\u003c/strong\u003e, 267\u0026ndash;277 (2020).\u003c/li\u003e\n\u003cli\u003eSchwing, R., Jocteur, E., Wein, A., No\u0026euml;, R. \u0026amp; Massen, J. J. M. Kea cooperate better with sharing affiliates. \u003cem\u003eAnim. Cogn. \u003c/em\u003e\u003cstrong\u003e19\u003c/strong\u003e, 1093\u0026ndash;1102 (2016).\u003c/li\u003e\n\u003cli\u003ePlotnik, J. M., Lair, R., Suphachoksahakun, W. \u0026amp; De Waal, F. B. M. Elephants know when they need a helping trunk in a cooperative task. \u003cem\u003eProc. Natl. Acad. Sci. USA \u003c/em\u003e\u003cstrong\u003e108\u003c/strong\u003e, 5116\u0026ndash;5121 (2011).\u003c/li\u003e\n\u003cli\u003eBalakhonov, D. \u0026amp; Rose, J. Crows rival monkeys in cognitive capacity. \u003cem\u003eSci. Rep. \u003c/em\u003e\u003cstrong\u003e7\u003c/strong\u003e, 1\u0026ndash;8 (2017).\u003c/li\u003e\n\u003cli\u003eLazareva, O. F. \u003cem\u003eet al.\u003c/em\u003e Transitive responding in hooded crows requires linearly ordered stimuli. \u003cem\u003eJ. Exp. Anal. Behav. \u003c/em\u003e\u003cstrong\u003e82\u003c/strong\u003e, 1\u0026ndash;19 (2004).\u003c/li\u003e\n\u003cli\u003eMagnotti, J. F., Katz, J. S., Wright, A. A. \u0026amp; Kelly, D. M. Superior abstract-concept learning by Clark\u0026rsquo;s nutcrackers (\u003cem\u003eNucifraga columbiana\u003c/em\u003e). \u003cem\u003eBiol. Lett. \u003c/em\u003e\u003cstrong\u003e11\u003c/strong\u003e, 20150148 (2015).\u003c/li\u003e\n\u003cli\u003eMagnotti, J. F., Wright, A. A., Leonard, K., Katz, J. S. \u0026amp; Kelly, D. M. Abstract-concept learning in Black-billed magpies (\u003cem\u003ePica hudsonia\u003c/em\u003e). \u003cem\u003ePsychon. Bull. Rev. \u003c/em\u003e\u003cstrong\u003e24\u003c/strong\u003e, 431\u0026ndash;435 (2017).\u003c/li\u003e\n\u003cli\u003eSamuleeva, M. V. \u0026amp; Smirnova, A. A. Emergence of reflexivity relation without identity matching-to-sample training in hooded crows (\u003cem\u003eCorvus cornix\u003c/em\u003e). \u003cem\u003eCogn. Neurosci. \u003c/em\u003e\u003cstrong\u003e65\u003c/strong\u003e, (2020).\u003c/li\u003e\n\u003cli\u003eSmirnova, A. A., Lazareva, O. F. \u0026amp; Zorina, Z. A. Use of number by crows: Investigation by matching and oddity learning. \u003cem\u003eJ. Exp. Anal. Behav. \u003c/em\u003e\u003cstrong\u003e73\u003c/strong\u003e, 163\u0026ndash;176 (2000).\u003c/li\u003e\n\u003cli\u003eSmirnova, A. A., Zorina, Z. A., Obozova, T. A. \u0026amp; Wasserman, E. A. Crows spontaneously exhibit analogical reasoning. \u003cem\u003eCurr. Biol. \u003c/em\u003e\u003cstrong\u003e25\u003c/strong\u003e, 256\u0026ndash;260 (2015).\u003c/li\u003e\n\u003cli\u003eSmirnova, A. A., Obozova, T. A., Zorina, Z. A. \u0026amp; Wasserman, E. A. How do crows and parrots come to spontaneously perceive relations-between-relations?. \u003cem\u003eCurr. Opin. Behav. Sci. \u003c/em\u003e\u003cstrong\u003e37\u003c/strong\u003e, 109\u0026ndash;117 (2021).\u003c/li\u003e\n\u003cli\u003eSeed, A. M., Emery, N. J. \u0026amp; Clayton, N. S. Intelligence in corvids and apes: A case of convergent evolution?. \u003cem\u003eEthology \u003c/em\u003e\u003cstrong\u003e115\u003c/strong\u003e, 401\u0026ndash;420 (2009).\u003c/li\u003e\n\u003cli\u003eWilson, B., Mackintosh, N. J. \u0026amp; Boakes, R. A. Transfer of relational rules in matching and oddity learning by pigeons and corvids. \u003cem\u003eQ. J. Exp. Psychol. \u003c/em\u003e\u003cstrong\u003e37\u003c/strong\u003e, 313\u0026ndash;332 (1985).\u003c/li\u003e\n\u003cli\u003eWright, A. A., Magnotti, J. F., Katz, J. S., Leonard, K. \u0026amp; Kelly, D. M. Concept learning set-size functions for Clark\u0026rsquo;s nutcrackers. \u003cem\u003eJ. Exp. Anal. Behav. \u003c/em\u003e\u003cstrong\u003e105\u003c/strong\u003e, 76\u0026ndash;84 (2016).\u003c/li\u003e\n\u003cli\u003eWright, A. A. \u003cem\u003eet al.\u003c/em\u003e Corvids outperform pigeons and primates in learning a basic concept. \u003cem\u003ePsych. Sci. \u003c/em\u003e\u003cstrong\u003e28\u003c/strong\u003e, 437\u0026ndash;444 (2017).\u003c/li\u003e\n\u003cli\u003eG\u0026uuml;nt\u0026uuml;rk\u0026uuml;n, O. \u0026amp; Bugnyar, T. Cognition without Cortex. \u003cem\u003eTrends Cogn. Sci. \u003c/em\u003e\u003cstrong\u003e20\u003c/strong\u003e, 291\u0026ndash;303 (2016).\u003c/li\u003e\n\u003cli\u003eG\u0026uuml;nt\u0026uuml;rk\u0026uuml;n, O., Str\u0026ouml;ckens, F., Scarf, D. \u0026amp; Colombo, M. Apes, feathered apes, and pigeons: differences and similarities. \u003cem\u003eCurr. Opin. Behav. Sci. \u003c/em\u003e\u003cstrong\u003e16\u003c/strong\u003e, 35\u0026ndash;40 (2017).\u003c/li\u003e\n\u003cli\u003eG\u0026uuml;nt\u0026uuml;rk\u0026uuml;n, O., Von Eugen, K., Packheiser, J. \u0026amp; Pusch, R. Avian pallial circuits and cognition: A comparison to mammals. \u003cem\u003eCurr. Opin. Neurobiol. \u003c/em\u003e\u003cstrong\u003e71\u003c/strong\u003e, 29\u0026ndash;36 (2021).\u003c/li\u003e\n\u003cli\u003eSayol, F., Lefebvre, L. \u0026amp; Sol, D. Relative brain size and its relation with the associative pallium in birds. \u003cem\u003eBrain Behav. Evol. \u003c/em\u003e\u003cstrong\u003e87\u003c/strong\u003e, 69\u0026ndash;77 (2016).\u003c/li\u003e\n\u003cli\u003eStr\u0026ouml;ckens, F. \u003cem\u003eet al.\u003c/em\u003e High associative neuron numbers could drive cognitive performance in corvid species. \u003cem\u003eJ. Comp. \u003c/em\u003e\u003cem\u003eNeurol. \u003c/em\u003e\u003cstrong\u003e530\u003c/strong\u003e, 1588\u0026ndash;1605 (2022).\u003c/li\u003e\n\u003cli\u003eVon Eugen, K., Tabrik, S., G\u0026uuml;nt\u0026uuml;rk\u0026uuml;n, O. \u0026amp; Str\u0026ouml;ckens, F. A comparative analysis of the dopaminergic innervation of the executive caudal nidopallium in pigeon, chicken, zebra finch, and carrion crow. \u003cem\u003eJ. Comp. \u003c/em\u003e\u003cem\u003eNeurol. \u003c/em\u003e\u003cstrong\u003e528\u003c/strong\u003e, 2929\u0026ndash;2955 (2020).\u003c/li\u003e\n\u003cli\u003eShumaker, R. W., Walkup, K. R. \u0026amp; Beck, B. B. \u003cem\u003eAnimal Tool Behavior: The Use and Manufacture of Tools by Animals\u003c/em\u003e (Johns Hopkins University Press, 2011). \u003c/li\u003e\n\u003cli\u003eDavenport, J., O\u0026rsquo;Callaghan, M. J. A., Davenport, J. L. \u0026amp; Kelly, T. C. Mussel dropping by Carrion and Hooded crows: biomechanical and energetic considerations. \u003cem\u003eJ. Field Ornithol. \u003c/em\u003e\u003cstrong\u003e85\u003c/strong\u003e, 196\u0026ndash;205 (2014).\u003c/li\u003e\n\u003cli\u003eHeinrich, B. Raven Tool Use?. \u003cem\u003eThe Condor \u003c/em\u003e\u003cstrong\u003e90\u003c/strong\u003e, 270\u0026ndash;271 (1988).\u003c/li\u003e\n\u003cli\u003eSmirnova, A. A., Bulgakova, L. R., Cheplakova, M. A. \u0026amp; Jelbert, S. A. Hooded crows (\u003cem\u003eCorvus cornix\u003c/em\u003e) manufacture objects relative to a mental template. \u003cem\u003eAnim. Cogn. \u003c/em\u003e\u003cstrong\u003e27\u003c/strong\u003e, 36 (2024).\u003c/li\u003e\n\u003cli\u003eJelbert, S. A., Hosking, R. J., Taylor, A. H. \u0026amp; Gray, R. D. Mental template matching is a potential cultural transmission mechanism for New Caledonian crow tool manufacturing traditions. \u003cem\u003eSci. Rep. \u003c/em\u003e\u003cstrong\u003e8\u003c/strong\u003e, 8956 (2018).\u003c/li\u003e\n\u003cli\u003eLaumer, I. B., Jelbert, S. A., Taylor, A. H., R\u0026ouml;ssler, T. \u0026amp; Auersperg, A. M. I. Object manufacture based on a memorized template: Goffin\u0026rsquo;s cockatoos attend to different model features. \u003cem\u003eAnim. Cogn. \u003c/em\u003e\u003cstrong\u003e24\u003c/strong\u003e, 457\u0026ndash;470 (2021).\u003c/li\u003e\n\u003cli\u003eBagotskaya, M. S., Smirnova, A. A. \u0026amp; Zorina, Z. A. \u003cem\u003eCorvidae\u003c/em\u003e can understand logical structure in baited string-pulling tasks. \u003cem\u003eNeurosci. Behav. Physi. \u003c/em\u003e\u003cstrong\u003e42\u003c/strong\u003e, 36\u0026ndash;42 (2012).\u003c/li\u003e\n\u003cli\u003eHofmann, M. M., Cheke, L. G. \u0026amp; Clayton, N. S. Western scrub-jays (\u003cem\u003eAphelocoma californica\u003c/em\u003e) solve multiple-string problems by the spatial relation of string and reward. \u003cem\u003eAnim. Cogn. \u003c/em\u003e\u003cstrong\u003e19\u003c/strong\u003e, 1103\u0026ndash;1114 (2016).\u003c/li\u003e\n\u003cli\u003eRiemer, S., M\u0026uuml;ller, C., Range, F. \u0026amp; Huber, L. Dogs (\u003cem\u003eCanis familiaris\u003c/em\u003e) can learn to attend to connectivity in string pulling tasks. \u003cem\u003eJ. Comp. Psychol.\u003c/em\u003e\u003cem\u003e \u003c/em\u003e\u003cstrong\u003e128\u003c/strong\u003e, 31\u0026ndash;39 (2014).\u003c/li\u003e\n\u003cli\u003eWakonig, B., Auersperg, A. M. I. \u0026amp; O\u0026rsquo;Hara, M. String-pulling in the Goffin\u0026rsquo;s cockatoo (\u003cem\u003eCacatua goffiniana\u003c/em\u003e). \u003cem\u003eLearn. Behav. \u003c/em\u003e\u003cstrong\u003e49\u003c/strong\u003e, 124\u0026ndash;136 (2021).\u003c/li\u003e\n\u003cli\u003eBaciadonna, L., Cornero, F. M., Clayton, N. S. \u0026amp; Emery, N. J. Mirror-mediated string-pulling task in Eurasian jays (\u003cem\u003eGarrulus glandarius\u003c/em\u003e). \u003cem\u003eAnim. Cogn. \u003c/em\u003e\u003cstrong\u003e25\u003c/strong\u003e, 691\u0026ndash;700 (2022).\u003c/li\u003e\n\u003cli\u003eAlem, S. \u003cem\u003eet al.\u003c/em\u003e Associative mechanisms allow for social learning and cultural transmission of string pulling in an insect. \u003cem\u003ePLoS Biol. \u003c/em\u003e\u003cstrong\u003e14\u003c/strong\u003e, e1002564 (2016).\u003c/li\u003e\n\u003cli\u003eDanel, S., Von Bayern, A. M. P. \u0026amp; Osiurak, F. Ground-hornbills (\u003cem\u003eBucorvus\u003c/em\u003e) show means-end understanding in a horizontal two-string discrimination task. \u003cem\u003eJ. Ethol. \u003c/em\u003e\u003cstrong\u003e37\u003c/strong\u003e, 117\u0026ndash;122 (2019).\u003c/li\u003e\n\u003cli\u003eGaycken, J., Picken, D. J., Pike, T. W., Burman, O. H. P. \u0026amp; Wilkinson, A. Mechanisms underlying string-pulling behaviour in green-winged macaws. \u003cem\u003eBehaviour \u003c/em\u003e\u003cstrong\u003e156\u003c/strong\u003e, 619\u0026ndash;631 (2019).\u003c/li\u003e\n\u003cli\u003eChaves Molina, A. B., Cullell, T. M. \u0026amp; Mim\u0026oacute;, M. C. String-pulling in African grey parrots (\u003cem\u003ePsittacus erithacus\u003c/em\u003e): performance in discrimination tasks. \u003cem\u003eBehaviour \u003c/em\u003e\u003cstrong\u003e156\u003c/strong\u003e, 847\u0026ndash;857 (2019).\u003c/li\u003e\n\u003cli\u003eSchrauf, C. \u0026amp; Call, J. Great apes use weight as a cue to find hidden food. \u003cem\u003eAm. J. Primatol. \u003c/em\u003e\u003cstrong\u003e73\u003c/strong\u003e, 323\u0026ndash;334 (2011).\u003c/li\u003e\n\u003cli\u003eSchmidt, G. F. \u0026amp; Cook, R. G. Mind the gap: Means\u0026ndash;end discrimination by pigeons. \u003cem\u003eAnim. Behav. \u003c/em\u003e\u003cstrong\u003e71\u003c/strong\u003e, 599\u0026ndash;608 (2006).\u003c/li\u003e\n\u003cli\u003eWilson, M. Six views of embodied cognition. \u003cem\u003ePsychon. Bull. Rev. \u003c/em\u003e\u003cstrong\u003e9\u003c/strong\u003e, 625\u0026ndash;636 (2002).\u003c/li\u003e\n\u003cli\u003eGibbs, R. W. \u003cem\u003eEmbodiment and Cognitive Science\u003c/em\u003e (Cambridge University Press, 2006).\u003c/li\u003e\n\u003cli\u003eHaddara, N. \u0026amp; Rahnev, D. The impact of feedback on perceptual decision-making and metacognition: Reduction in bias but no change in sensitivity. \u003cem\u003ePsychol. Sci.\u003c/em\u003e \u003cstrong\u003e33\u003c/strong\u003e, 259\u0026ndash;275 (2022).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"physical cognition, problem-solving, causal reasoning, cooperation, loose-string paradigm, corvids","lastPublishedDoi":"10.21203/rs.3.rs-5405438/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5405438/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eIn a loose-string task an out-of-reach tray baited with food can only be retrieved by simultaneously pulling on both ends of a string threaded through the loops on the tray. This task is used to assess an animal's ability to cooperate, with each animal only having access to one end of the string. Some studies use the loose-string task in a pre-training phase, during which animals are individually taught to pull both ends of the string. Usually, no additional tests are conducted to determine whether the animals have understood how the loose string works. It is conceivable that a lack of knowledge of the causal basis of the loose-string task could make it more challenging to grasp how the partner can assist with it. Here, we tested whether Hooded crows could acquire some knowledge of the causal basis of the loose-string task. Prior to the critical test (Experiment 3), the birds were presented with two different tasks (Experiment 1 and 2) to allow them to acquire some knowledge of the causal basis of the task. The results may indicate that, as a consequence of the experience gained, some crows may have begun to understand how the loose string works.\u003c/p\u003e","manuscriptTitle":"Do Hooded crows (Corvus cornix) understand how a loose string works? ","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-12-16 08:14:56","doi":"10.21203/rs.3.rs-5405438/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-03-05T05:54:58+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-03-03T10:46:56+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"85615870724671056836282532609317047809","date":"2025-02-18T13:45:12+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-12-17T13:38:30+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"323126830236196196501690206177854033229","date":"2024-11-26T13:05:19+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"255643945562230933463273062819014392623","date":"2024-11-23T14:12:15+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-11-21T12:55:39+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-11-21T12:52:34+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2024-11-21T12:28:40+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-11-21T04:30:51+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2024-11-06T21:27:55+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"84bbdf2d-6f62-499d-9c55-d105963b10c8","owner":[],"postedDate":"December 16th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":40780908,"name":"Biological sciences/Neuroscience/Cognitive neuroscience/Problem solving"},{"id":40780909,"name":"Biological sciences/Neuroscience/Cognitive neuroscience/Intelligence"}],"tags":[],"updatedAt":"2025-05-05T15:58:28+00:00","versionOfRecord":{"articleIdentity":"rs-5405438","link":"https://doi.org/10.1038/s41598-025-99778-z","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2025-05-04 15:56:59","publishedOnDateReadable":"May 4th, 2025"},"versionCreatedAt":"2024-12-16 08:14:56","video":"","vorDoi":"10.1038/s41598-025-99778-z","vorDoiUrl":"https://doi.org/10.1038/s41598-025-99778-z","workflowStages":[]},"version":"v1","identity":"rs-5405438","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5405438","identity":"rs-5405438","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}