Human lateral occipital complex is invariant for position and size transformations at the single-neuron level

Human lateral occipital complex is invariant for position and size transformations at the single-neuron level | 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 Human lateral occipital complex is invariant for position and size transformations at the single-neuron level Vanhoyland Michaël, Janssen Peter, Theys Tom This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8222241/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 12 Mar, 2026 Read the published version in Scientific Reports → Version 1 posted 10 You are reading this latest preprint version Abstract Object recognition relies on neural invariance to changes in position and size, yet the underlying mechanisms in humans remain uncertain due to the challenges that accompany single-unit recordings in cortical areas. We conducted the first single-unit recordings investigating position and size invariance in the human lateral occipital complex (LO) using two microelectrode arrays implanted in distinct LO subregions of a single patient. Our findings reveal that LO neurons exhibit strong shape tuning and have smaller receptive fields than previously reported. Time-resolved correlational analyses and multidimensional scaling confirmed robust position invariance and size invariance for changes up to 2 octaves, mirroring properties of the macaque inferotemporal cortex. These results advance our understanding of LO’s critical role in visual processing and object recognition. Biological sciences/Biological techniques Biological sciences/Neuroscience Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Significance statement Macaques have served as a valuable animal model to better understand (human) brain functions across various regions, including shape selectivity and tolerance to object manipulations. However, the extent to which these findings apply to humans remains unclear. While non-invasive recording techniques can address some questions, studying shape selectivity requires measuring responses from individual neurons, which necessitates invasive intracortical recordings. Traditional depth electrodes, which target medial brain structures, are ill-suited for recording single neurons in the lateral occipital complex (LO), which is located on the brain's convexity. To overcome this, we implanted two micro-electrode arrays in different LO subregions and demonstrated that individual neurons exhibit tolerance to changes in position and size, mirroring patterns observed in the macaque inferotemporal cortex. Introduction A critical aspect of object recognition is invariance to stimulus transformations such as changes in location and size. For instance, consider a table with chairs around it: with a single glance, you can effortlessly categorize all chairs as "chairs" despite their varying positions relative to your viewpoint. Neural mechanisms underlying invariances have been extensively studied in monkeys across the ventral (notably areas TE and TEO in the inferotemporal cortex (ITC)) and the dorsal visual stream (lateral intraparietal cortex, LIP). 1 – 8 Macaque ITC, which shares many similarities with the human lateral occipital complex (LO) with regard to shape processing, 9–19 maintains stimulus selectivity across the receptive field and tolerates size modulations at the single-neuron level. 1 , 4 , 5 Extrapolating these findings to humans has been hindered by the need for single-neuron recordings to accurately investigate invariances. Activity in human LO, a pivotal region in the ventral stream for shape and object recognition, 18,20 undergoes adaptation in fMRI when a stimulus is repeatedly shown while simultaneously being changed in size or position, suggesting response invariance at the population level. 21 LO displays strong shape selectivity at the single-neuron level and features large receptive fields that can span contra-, ipsi-, or bilateral regions. 13 , 22 Within LO, patches of face- and body-selective cells have been identified, 12–14 though category responses appear primarily driven by underlying shape features. 12 , 15 Recently, we investigated two such invariances in LO and found that body-selectivity generalized for silhouettes and showed partial tolerance for rotational manipulations. 12 Other invariances, however, have thus far not been investigated. To address this, we recorded single-unit activity from two microelectrode arrays implanted in distinct LO subregions in the same patient. Our findings confirm the heterogeneous nature of LO receptive fields while revealing that these fields are smaller than previously reported. 13 Employing time-resolved correlational analyses and multidimensional scaling (MDS), we demonstrate that stimulus selectivity of a population of LO neurons is invariant to positional shifts within receptive fields and show partial tolerance to size manipulations. Results Receptive field mapping We recorded 28 visually responsive neurons using 2 multi-electrode arrays ( \(\:{N}_{A1}=13\:;{N}_{A2}=\:15)\) implanted in shape-selective lateral occipital cortex (Fig. 1 ). Sensitivity to image scrambling in both arrays was demonstrated in previous research, thereby confirming that both arrays were located in LO. 15 Moreover, array 2 was implanted in a body patch. 12 We mapped the central 28° × 28° receptive field by presenting a headless torso at 49 non-overlapping positions. For A1, the average receptive field was (para)foveal along the vertical meridian, while for A2, it was highly eccentric in the contralateral (right), inferior quadrant (Fig. 2 A). An overview of the number of neurons responding in each quadrant is presented in Table 1 . In A1, 67% of neurons responded foveally, compared to only 1 neuron (8%) in A2 (Fig. 2 B). Both arrays showed a contralateral bias, with 60% of neurons in A1 and 92% in A2 responding contralaterally. The average receptive field per array, including the positions of the receptive field centers per single-unit, are presented in Fig. 2 A and 2 B. The average size of the receptive field was 70 deg 2 and 98 deg 2 in A1 and A2, respectively. Table 1 Number of single-units of which the receptive field was mapped, and the position of the receptive field. Array 1 Array 2 N % N % Responsive 15 100 13 100 Foveal 10 67 1 8 Laterality Ipsilateral 3 20 0 0 Contralateral 9 60 12 92 Bilateral 1 7 1 8 Verticality Upper hemifield 3 20 1 8 Lower hemifield 4 27 6 46 Both 8 53 5 38 Shape selectivity Given the differences in stimulus selectivity and receptive fields across the two arrays, we presented distinct stimulus sets at different positions within the average receptive field, as displayed in Fig. 1 B. For A1, we presented 64 simple shapes (presented in Figure S1 ) at two positions along the vertical meridian (above and below the fixation marker). For array 2, we displayed 10 torso images at five locations in the contralateral lower quadrant. To ensure stable fixation, trials were automatically aborted if the gaze deviated beyond a predefined 3° × 3° fixation window. Furthermore, we verified that the average change in eye position during stimulus presentation (from onset to offset) was similar for different positions (Table S1 -2). This verification was particularly critical for array 1, where stimuli were presented closer to the fixation marker, as saccades toward the stimulus locations could compromise data interpretation. We recorded 48 neurons that were responsive in at least 2 locations ( \(\:{N}_{A1}=22\:;{N}_{A2}=26).\) For each neuron, we determined the preferred stimulus location (see Methods for criteria) and sorted the different stimuli according to the responses at the preferred location. As a measure of shape selectivity, we calculated the shape tuning index and depth of selectivity (see Methods). 6 , 23 In short, the shape tuning index (STI) measures how broadly a neuronal response is tuned for shapes by taking the net response to the best and worst stimulus into account, where STI = 1 signifies very strong tuning, and STI = 0 indicates no selectivity. In contrast, the selectivity depth takes all responses to all stimuli into account and is 1 if a neuron responds to only 1 stimulus, and 0 if it responds to all stimuli equally. The average STI in arrays 1 and 2 was 1.15 \(\:\pm\:\) 0.03 and 0.93 \(\:\pm\:\) 0.03, respectively. (Values >1 are caused by inhibitory responses to the least preferred stimulus.) The mean selectivity depth measured 0.79 \(\:\pm\:\) 0.02 and 0.67 \(\:\pm\:\) 0.03, respectively. The distribution of STI and selectivity depth are presented in Fig. 3 B. Both indices point towards strong shape tuning, which is further visualized in Fig. 3 A where the average response to the worst stimulus is negligible, while the preferred stimulus evoked on average a net increase of 25–40 spikes/s. Position invariance of the selectivity within the receptive field To investigate whether the shape preference is maintained within the receptive field, the stimuli at the other, non-preferred location(s) were ranked according to the preference at the preferred location. An example single unit with strong shape selectivity (STI = 1.05; selectivity depth = 0.87) and strong selectivity invariance across the receptive field is presented in Fig. 4 . For all shape-selective neurons (N = 48), we calculated the Pearson correlation coefficient ‘r’ between the ranked responses at the preferred and non-preferred locations over time after stimulus onset. In case of selectivity invariance at the population level, a significant correlation exists between preferred and non-preferred locations. For both arrays, the ranked responses in the preferred and non-preferred locations were strongly correlated starting from 90 ms and 80 ms (middle of 100 ms bin) in array 1 and 2, respectively (peak correlation in array 1: r = 0.32, p = 3.62 x 10 –34 ; array 2: r = 0.57, p = 1.07 x 10 –83 , two-sided t-test, Fig. 5 A). The lower correlations obtained in array 1 are partially due to the larger number of used shapes. We therefore repeated the analysis with 10 stimuli (5 best and 5 worst per neuron) to match the number of stimuli in array 2, which resulted in correlations similar to array 2 (r = 0.53, 2.95 x 10 –19 , two-sided t-test, Figure S3). Because responses are variable between trials of the same stimulus, we also calculated the intertrial correlations (see Methods), which serves as theoretical maximal obtainable correlation per array. For both arrays, the intertrial correlations exceeded the correlations between preferred and non-preferred location (peak correlation in array 1: r = 0.52 p = 5.54 x 10 –94 ; array 2: r = 0.73; p = 4.69 x 10 –57 , two-sided t-test, Fig. 5 A). For each neuron, we fitted the ordinary least squares (OLS) linear regression through the ranked response in the non-preferred locations (Fig. 5 C). In case of selectivity invariance within the receptive field, the slope of the regression should be significantly different from 0 (e.g. negative slope in case of descending ranking). For both arrays, the average slope was strictly negative (Array 1: \(\:\beta\:\) = -0.454 \(\:\pm\:\:0.075\) , Array 2: \(\:\beta\:\) = -0.627 \(\:\pm\:\) 0.050). Lastly, we performed multidimensional scaling to visualize, in an unbiased way, if a stimulus would be grouped together despite being presented in different locations on the screen (Fig. 6 ). The position in which a stimulus was presented on the screen largely influenced the position on the two-dimensional representation, which is a direct result of receptive field interactions. However, the distance in the two-dimensional space was smaller between identical shapes presented at different positions compared to different shapes presented at different positions, signifying selectivity invariance (Fig. 6 B). Size invariance We presented 20 torsos in 3 different sizes (2, 4 and 8 visual degrees) in the receptive field (contralateral lower quadrant) and recorded 29 visually responsive single units on array 2, of which 14 were responsive to \(\:\ge\:\) 1 size (14 neurons were only responsive to the 8° torsos). Ten neurons were only responsive to 4° and 8°, 1 neuron to only 2° and 8° and 3 neurons to all 3 sizes. The responses from a neuron responsive to all 3 sizes is presented in Fig. 7 A. Similar to the analysis done for position invariance, we ranked the responses to different sized torso’s according to sorting order obtained from the size eliciting the strongest response. Subsequently, we calculated the time-resolved Pearson correlation (Fig. 7 B-C) and the slope from the ordinary least squares linear regression (Fig. 7 D). Peak correlation was obtained in the 270–370 ms interval (r = 0.36, p = 1.28 x 10 –11 , two-sided t-test). Because this analysis pooled all size pairs (2° vs 4°, 2° vs 8° and 4° vs 8°), we also calculated the correlation between each pair individually to exclude that the main effect was carried by the correlations from only 1 pair. As visualized in Fig. 7 C, we obtained significant correlations between the sorted responses for all pairs (2° vs 4°: r = 0.29, p = 1.13 x 10 − 9 , 2° vs 8°: r = 0.28, p = 2.12 x 10 − 5 ; 4° vs 8°: r = 0.40, p = 3.84 x 10 –11 ; two-sided t-test). Finally, also the slope from the OLS linear regression was significantly negative ( \(\:\beta\:\) = -0.441 \(\:\pm\:\) 0.076), indicating size invariance at the population level. Discussion We conducted the first single-unit recordings in the human lateral occipital complex (LO) to investigate neuronal response invariance to positional shifts and size variations within the receptive field. Our findings demonstrate that LO neurons exhibit strong shapes tuning and robust tolerance to positional changes. Shape selectivity was preserved across size variations of up to two octaves (fourfold size increase). The invariance of the shape selectivity across stimulus transformations highlights LO’s crucial role in object recognition. Due to the semi-acute nature of our multielectrode recordings in a human subject, mapping the precise receptive field for each neuron individually with its optimal stimulus was not feasible. Instead, we estimated the average receptive field, tested various shapes at different positions within this field, and presented shapes of varying sizes at the center of the average receptive field to maximize the number of responsive units. Non-responsive positions and sizes were excluded from pairwise analysis, as they were outside the receptive field. Notably, nearly all neurons (28 of 29) responded to 8° × 8° shapes, while only 4 of 29 responded to 2° × 2° shapes. This discrepancy may suggest suboptimal centering within the receptive field, or that the smaller shapes were too small to reliably evoke responses. Additionally, the 8° × 8° shapes may have extended beyond some receptive field boundaries. Estimated average receptive field sizes were 70 deg² and 98 deg² for array 1 and 2, respectively. However, since some neurons in array 2 responded to the most eccentric positions as visualized in Fig. 2 C (+ 14° on the x-axis, and − 14° on the y-axis), the true receptive field size for array 2 might be underestimated. Our estimates are notably smaller than a prior study’s reported 473 deg². 13 This could possibly be explained by differences in used stimuli (4° × 4° torsos here vs. 10° × 6° shapes in Decramer & Premereur et al. 13 ), mapped area (28° × 28° vs 50° × 30°), or inherent heterogeneity within LO and across subjects. Both arrays showed contralateral receptive field biases, consistent with fMRI findings. 22 , 24 – 28 Although limited to a single patient, our data provides robust evidence for size and positional invariance, the latter consistent across 2 distinct subregions of LO with different response profiles. This shows that these invariances are widespread within LO. Array 2 was positioned in a body patch, 29 as identified in previous work (P1 in Ramirez & Vanhoyland et al. 12 ), and was tested for position and size invariance using headless torso stimuli. These contained detailed features, such as hands, unlike the simpler silhouettes used for array 1. Although these details could theoretically drive responses independently of shape contours, this seems unlikely, as a linear decoder trained on these torsos generalized to their silhouette counterparts (as shown in Ramirez & Vanhoyland et al. 12 ), which is also consistent with macaque body-patch recordings. 30 Non-human primates have been extensively used to evaluate response invariance in both ventral and dorsal visual stream at the single-unit level. 1 – 8 , 31 , 32 The inferotemporal cortex (ITC) in the ventral stream maintains shape selectivity despite positional shifts and exhibits robust size invariance. 1 , 2 Despite maintenance of the ranking order across transformations, the absolute magnitude of responses might change, similar to our observations. 7 , 8 However, the response pattern of multiple single-neurons is necessary in ITC to decode objects across changes in size and position. 31 Recent advancements in intracortical recordings in the human lateral occipital (LO) cortex reveal striking parallels with macaque ITC, such as similarities in coding of shape, stereopsis, bodies and faces. 9 – 17 Our findings corroborate these similarities, showing that human LO maintains stimulus selectivity across size changes of up to two octaves (the maximum that was tested), compared to tolerance of up to 4 octaves in macaque ITC. 1 , 2 Time-resolved correlational analysis revealed significant Pearson correlation coefficients of 0.29 for 2° vs. 4° (1 octave), 0.28 for 2° vs. 8° (2 octaves), and 0.40 for 4° vs. 8° (1 octave). The lower correlations for comparisons involving 2° stimuli likely stem from fewer neurons responding to these smaller shapes. Single-neuron recordings in human LO are lacking, and current evidence is based on fMRI and EEG. LO activity remains invariant to stimulus size changes of up to fourfold in fMRI, 18,21 and most studies support invariant stimulus representation across the visual field. 21 , 24 , 33 – 35 BOLD signal adaptation occurs when stimuli are repeatedly presented with varying size or position, suggesting invariant neural representations, as intolerance would lead to activation of distinct neural populations, preventing BOLD suppression. 21 Additionally, both individual exemplars and superordinate categories are decodable across the visual field, even between hemifields. 28 , 36 , 37 It was hypothesizes that such location-tolerant coding could be explained by large receptive fields crossing the meridians, 28 based on the receptive fields from macaque ITC that typically range between 5 and 30°. 2,7,38–40 Our data, in line with Decramer & Premereur et al. 13 , supports this theory since a considerable number of neurons crossed the vertical or horizontal meridian. In contrast, one fMRI study found no evidence for individuation (ability to differentiate between individual objects based on the spatial response patterns) for objects across different positions within and between same hemifields. 25 Notably, position-invariant category information extends beyond LO to most category-selective regions such as the extrastriate body area and fusiform face area. 35 Yet, LO appears uniquely capable of distinguishing stimulus location on cluttered backgrounds, with EEG recordings indicating a latency of 140–200 ms. 37 While fMRI adaptation paradigms aim to be less susceptible to the obscuring effects of averaging neural activity of potentially selective neural populations, all fMRI and EEG signals remain an ensemble of thousands of neurons simultaneously. Our single-neuron data therefore provide crucial validation of these neuroimaging findings, confirming that stimulus selectivity in human LO persists across size and position transformations. Moreover, the high temporal resolution of our recordings reveals that these invariances emerge during feedforward propagation, independent of top-down influences. These findings bridge a critical gap between non-human primate and human studies, offering direct evidence of single-neuron invariance in human LO. Methods Study approval We obtained ethical approval (study number s53126) for conducting semi-chronic microelectrode recordings using the Utah array in patients undergoing invasive epilepsy monitoring. The study protocol was approved by the local ethical committee (Ethische Commissie Onderzoek UZ/KU Leuven). The study was carried out in accordance with the principles of the Declaration of Helsinki, the principles of good clinical practice, and all applicable regulatory requirements set by the Federal Agency for Medicines and Health Products (FAGG). To ensure safety and compliance, strict adherence to imposed safety measures was maintained. This included the use of case report forms and detailed reports on any (serious) adverse events. All data collected during the study were encrypted and stored at the University Hospitals Leuven. The study was thoroughly discussed with the patient during the pre-operative consultation, which took place at least six weeks before the surgery. The patient was informed about the risks associated with micro-electrode array implantation, such as infection and hemorrhage. No additional incisions were made for the purpose of this study. Written informed consent was obtained from the patient on the evening prior to the surgery. Clinical information We obtained invasive intracranial recordings from 2 arrays from the left lateral occipital cortex in a patient with refractory epilepsy. She was a 58-year old female suffering from recurrent psychotic episodes. She was taking chronic anti-seizure medications (ASM) during admission: Levetiracetam 2000 mg, Phenytoin 100 mg, Clonazepam 1 mg and Lacosamide 300 mg, all twice daily. During admission to the epilepsy ward, the ASM was adjusted based on the number of detected seizures. Target locations of intracranial electrodes were determined by the epileptologist in a multidisciplinary consensus meeting and were based on electroclinical findings and non-invasive multimodal imaging. Multi-electrode arrays were implanted via the craniotomy performed for the implantation of a subdural grid, adjacent to the grid (Figure S4). The arrays were removed during the same surgery for subdural grid removal, 14 days after implantation. Both insertion and removal of electrodes were done under general anesthesia. No new neurological deficits, nor implantation-related complications (infection, hemorrhage, wound problems) were recorded (Figure S5). Although the presumed epileptogenic zone encompassed the lateral occipital cortex based on multimodal preoperative assessment, no focal onset could be identified. Therefore, she was deemed not to be a good candidate for epilepsy surgery. Invasive recordings We recorded from two 96-channel arrays (Utah Array™ – Blackrock Neurotech) from day 1 until the day before electrode removal. Arrays were inserted using a pneumatic inserter wand, provided by Blackrock Neurotech, with a single hit (20 psi). We closed the dura above the array and placed the bone flap on top to secure the array and prevent array flotation. Reference wires were placed subdurally and ground wires epidurally. The signal was digitally amplified using a Cereplex M head stage from Blackrock Neurotech, connected to a 128-channel neural signal processor (NeuroPort), and sampled using Central Software at a rate of 30 kHz. We applied a 750 Hz high-pass filter for spiking activity. The multi-unit detection threshold was placed at -3 standard deviations of the signal. The variable signal quality in the immediate postoperative period resulted in a difference in the number of responsive and selective channels during each recording session. Single units were manually extracted offline using Plexon Offline Sorter v4.7.1. Imaging T1-weighted images were acquired using a 3-T MR scanner (Achieva dstream, Philips Medical Systems, Best, the Netherlands) at the University Hospitals Leuven for presurgical planning. The imaging parameters were as follows: time of echo 3.1 ms, inversion time 900 ms, slice thickness 1.0 mm, 400 × 400 matrix size, and 283 coronal slices. Following electrode placement, a CT scan (Siemens) was performed with a slice thickness of 1 mm, a voltage of 120 kV, and a dose length product of 819 mGy.cm. The CT scan was used to identify the exact electrode locations and rule out any hemorrhage. This postoperative CT was later fused with a postoperative MRI performed months after surgery to rule out structural alterations due to electrode insertion. No structural alterations (gliosis, ischemia, hemosiderin) were seen at the implantation sites (Figure S4). 15 Imaging preprocessing was done with SPM12 software (Wellcome Department of Cognitive Neurology, London, UK) running on MATLAB (Mathworks, Natick, MA, USA). The preprocessing steps included (1) setting the anterior commissure as the origin for both the MRI and CT scan (2) realigning the images, (3) co-registering the MRI and CT images and (4) warping the MRI and coregistered CT to a common high-resolution brain atlas (MNI152). Freesurfer was used to generate cortical 3D renderings of the subjects' pial surfaces. 41 The precise electrode location on the cortical surface and the MNI coordinates were determined using iElectrodes Software. 42 Stimulus presentation We conducted experiments in a dimmed hospital room using custom-built software for stimulus presentation. Stimuli were presented on a 60 Hz DELL-P2418HZM LED monitor, while the patient sat 60 cm away from the screen (1 pixel = 0.026°). Fixation was monitored using continuous eye-movement tracking (left or right eye, 1000 Hz, Eyelink 1000 plus in head free-to move mode). The patient was instructed to focus on a small red square (0.2 × 0.2°) positioned at the center of the display. If the gaze deviated from the electronically preset 3° by 3° fixation window, the trial was aborted. To synchronize the data, a photodiode was attached to the upper left corner of the screen, which detected a bright flash coinciding with the first frame of the stimulus. This flash was invisible to the patient. To accurately record baseline spike rate activity, an intertrial interval of at least 100 ms was introduced (i.e. the offset of the stimulus and the onset of fixation for the next trial). Stimulus sets and experimental designs Receptive field We initially presented 10 images of headless torso’s 43,44 (size 4° by 4°) at the fixation marker while the patient passively fixated, and selected the torso that elicited the strongest response across all neurons from array 2, designating this as the "preferred body" (Figure S2). Stimulus presentation lasted 250 ms, and each trial was preceded by 300 ms of fixation. This was done on day 3 after implantation. This preferred body was then displayed at 49 (7 by 7) non-overlapping positions, thereby mapping the central 28° x 28°. Locations were mapped in randomized order. The patient performed a distractor task, pressing a button with her right hand whenever a cross appeared on the fixation marker, which occurred in approximately 2% of trials. Stimulus presentation lasted 250 ms, and each trial was preceded by 300 ms of fixation. Due to time constraints and the multi-electrode recording technique, we could not test the receptive field of each neuron individually with its optimal stimulus. The experiment was performed on day 12 after implantation. Invariance for position within receptive field We used two stimulus sets to test for invariance due to the different selectivity profiles observed in both arrays. The stimuli were presented in different non-overlapping locations within the average receptive field for a particular array. Individual stimuli and on-screen location were presented in a randomized order. The first set consisted of 64 white shapes presented on a black background, of which 30% was filled with random black noise. Shapes were approximately 4° by 4°. All 64 shapes were presented 4° above or below the fixation marker (2 locations) with its center on the vertical meridian, thereby not crossing the horizontal meridian. Each shape was presented \(\:\pm\:\) 10 times in each location. Stimulus presentation lasted 250 ms, and each trial was preceded by 300 ms of fixation. This set was recorded in array 1 on day 11. The second set consisted of the same 10 torso’s as described for the receptive field testing (4° by 4°). All ten torsos were presented \(\:\pm\:\) 20 times in 5 locations in the lower right quadrant on a white background. The x- and y- coordinates of the center locations were: (4°,-8°),(8°,-8°), (12°,-8°), (8°,-12°), (12°,-12°). Stimulus presentation lasted 250 ms, and each trial was preceded by 300 ms of fixation. This set was recorded in array 2 on day 6. In both experiments, the patient performed a distractor task by pressing a button when a cross was presented on the fixation marker ( \(\:\pm\:\) 2% of trials). Invariance for changes in size This experiment was only done for array 2. We presented 20 torso’s (10 of which were used for the position invariance experiment) in 3 different sizes (2° by 2°, 4° by 4° and 8° by 8°). All torsos were presented \(\:\pm\:\) 8 times in all 3 sizes on a white background. The torso’s were presented within the receptive field in the right lower quadrant with center location x = 10° and y = -10°. Stimulus presentation lasted 200 ms, and each trial was preceded by 300 ms of fixation. The patient was passively fixating. This set was recorded on day 4. Analysis Data analysis We analyzed data using custom-written MATLAB R2023b (Mathworks, Natick, MA, USA) scripts. Net spike firing rates were calculated from raw multi-unit and single-unit firing rates by subtracting baseline activity (250 or 300 ms interval before stimulus onset) from the entire trial, binned in 10 ms intervals. Analyses were restricted to visually responsive sites. To assess the responsiveness of each recorded site, we performed two-sided dependent samples t-tests comparing spike rates during baseline and response windows for each stimulus category, condition or position separately. This approach ensured that sites with a strong preference for a specific category or stimulus are not overlooked by a broad test across all stimuli. For each experiment, a site was considered visually responsive if any condition showed a significant difference from baseline (p < 0.005 to account for multiple testing) in the 150 ms window following the response latency, the ‘response window’ (Array 1: 130–280 ms; Array 2: 200–350 ms). Response windows have been reported in a previous research article (array 1 corresponds to array 3, and array 2 to array 4). 45 Data normalization All subsequent analyses utilized baseline-subtracted average net firing rates or z-normalized net firing rates. Z-normalization was performed for each 10 ms-binned multi-unit or single-unit recording as follows: $$\:Z\left(i\right)=\:\frac{R\left(i\right)-{R}_{mean}}{{R}_{std}}$$ where \(\:Z\left(i\right)\) corresponds to the Z-normalized net spike rate for a specific 10ms bin, \(\:R\left(i\right)\) to the average net response from the same 10 ms bin, \(\:{R}_{mean}\) is the average net spike rate over all trials in the response window, and \(\:{R}_{std}\) is the standard deviation in the response window. Within the experiments testing for invariance, \(\:{R}_{mean}\) and \(\:{R}_{std}\) were taken from all stimuli shown in the same position (for position invariance) or same size (for size invariance). This method ensures that the firing rates are standardized, allowing for consistent comparisons across different conditions and sites. Receptive field We calculated the average response across all 49 locations within the response window, selecting only units that were responsive in at least one location with a minimum net spike rate of 5 spikes/s (p < 0.005, unpaired t-test comparing the response window to baseline, corrected for multiple testing). The location with the highest net response was identified as the center of the receptive field. Other locations were included in the receptive field if their response exceeded 50% of the center response and was significantly higher than baseline (p < 0.05, one-sided dependent samples t-test). The receptive field size was determined as the number of locations with a response ≥ 50% of the center response. To assess laterality (contralateral, ipsilateral, or bilateral) and verticality (upper hemifield, lower hemifield, or both), locations on the vertical and horizontal meridians were excluded, as these spanned the meridians. To generate an average receptive field per array, the net responses at each location were scaled per unit by dividing by the center response (thereby, center response equals 1). An average was then computed across all responsive units. For visualization, a Gaussian filter (σ = 0.5) was applied, followed by 2D cubic interpolation (refinement factor = 8). Shape-tuning index and depth of selectivity Both metrics were calculated on the position invariance experiment within the ‘preferred location’ (further explained in the Pearson correlation analysis) for all visually responsive neurons. The shape-tuning index quantifies the degree of shape tuning and is defined as \(\:\frac{MAX\:-\:MIN}{MAX}\) , in which ‘ MAX’ and ‘MIN’ are the net spike rate within the response window to the most and least preferred stimulus at the ‘preferred’ location, respectively. A value of 0 indicates no shape-tuning, whereas a value of 1 (or even > 1) indicates strong shape-tuning. The depth of selectivity was calculated as \(\:\frac{N-\:\frac{SUM}{MAX}}{N-1}\) , where N = number of stimuli presented at the preferred location, SUM = sum of the average net spike rate within the response window to all stimuli, and MAX = net spike within the response window to the preferred stimulus at the preferred location. The depth of selectivity is 1 if the neuron responds to only 1 stimulus, and 0 if it responds to all stimuli equally. 6 Pearson correlation analysis To quantify neural information in a time-resolved manner, firing rates were computed within 100 ms bins using a 10 ms sliding window. For each visually responsive single unit, the net response to each stimulus at each position (or size) was calculated within each time bin. Single-units that were only responsive in \(\:\ge\:\:\) 1 location for any stimulus (p < 0.005, dependent samples t-test comparing response window vs. baseline) were excluded from the analysis. For each single unit, the position (or size) eliciting the strongest response to any stimulus within each time bin was identified as the "preferred location/size" for that particular time bin. Stimuli were then ranked according to their z-normalized responses at the preferred location/size. For the 64 shapes presented in array 1, besides using all stimuli, we also used the 5 best and 5 worst stimuli only to make it more comparable to the 10 torso’s from array 2. We first estimated the noise ceiling, the maximal correlation that is theoretically obtainable. For this, we randomly split for each neuron the trials belonging to the preferred stimuli (which could be different for each single unit) at the preferred location/size in 2 equally sized groups. The average of both groups was taken, creating response "pairs" as a measure of intertrial variability ( \(\:{N}_{pairs\:}={N}_{single\:units\:}*{N}_{stimuli})\) . This process was iterated 1000 times for each time bin, and for each iteration we calculated the Pearson correlation coefficient (r) over all response pairs using the corrcoef function. Pearson’s r was averaged, and a 95% confidence interval was obtained from the null distribution of Pearson coefficients. To estimate the response invariance within the receptive field, the z-normalized responses from all other visually responsive positions/sizes (in which \(\:\ge\:\:\) 1 stimulus was visually responsive, p < 0.05, two-sided dependent samples t-test comparing response window vs. baseline), termed the "non-preferred positions/sizes", were sorted according to the ranking obtained from the preferred position/size, creating response "pairs" between the preferred and non-preferred positions/sizes. These response pairs, pooled across all visually responsive neurons, were used to compute Pearson’s r for each time bin. We performed Fisher’s z-transformation to derive 95% confidence intervals [ \(\:{r}_{lower\:boundary}\:{r}_{upper\:boundary}]\) of the Pearson coefficient r as follows: $$\:{Z}_{lower\:boundary}^{upper\:boundary}=0.5*\text{ln}\left(\frac{1+r}{1-r}\right)\pm\:\left(1.96*\frac{1}{\sqrt{N-3}}\right)\:\:with\:N={N}_{single\:units}*{N}_{stimuli}$$ $$\:{r}_{lower\:boundary}^{upper\:boundary}=\:\frac{{e}^{2Z\:}-1}{{e}^{2Z}+1}\:wih\:Z={Z}_{upper\:boundary}\:or\:{Z}_{lower\:boundary}$$ Significance was assessed with the associated two-tailed t-test of the null hypothesis that the population correlation coefficient equals zero (df = n – 2). Ordinary Least Squares (OLS) regression analysis We fitted a linear regression through the sorted (from high to low) responses, per single unit, from the non-preferred positions/sizes using the Ordinary Least Squares method. We calculated the average slope and 95% confidence intervals. A negative regression slope ( \(\:\beta\:\) ) significantly different from zero signifies response invariance to some degree between different locations. The reported slope equals the calculated slope, multiplied by the number of stimuli (64 for array 1; 10 for array 2) to be able to compare slopes across arrays. Multi-dimensional scaling To visualize the similarity relationships in the neuronal population between the torso’s/shapes and the position within the receptive field, we employed multidimensional scaling (MDS). The neuronal distances were computed as the average Euclidean distance of the baseline-subtracted average Z-normalized firing rates across neurons for each image pair using MATLAB’s pdist function. This way we generated a dissimilarity matrix, with both dimensions corresponding to the number of individual images, that held the Euclidian distances measured in multidimensional space ( \(\:{N}_{dimensions\:}=\:\) number of visually responsive single-units). The MDS procedure using MATLAB’s mdscale function was applied to the dissimilarity matrix of each experiment. To evaluate the alignment between the two-dimensional embedding and the actual multi-dimensional neuronal distances, we calculated the Pearson correlation between them. The resulting correlation coefficient is displayed in each multidimensional scaling plot. High correlation values indicate that the multidimensional scaling plots provide a reasonable approximation of the underlying similarity relationships. Given that we observed high correlations, we calculated the average distance (in this two-dimensional embedding) ‘within’ and ‘between’ groups in 100ms bins with a 10ms sliding window (e.g. ‘ within’ a certain location in the receptive field and ‘ between ’ different locations). To calculate the average Euclidian distance, we determined, for each stimulus individually, the distance to all stimuli within the same group ( \(\:{N}_{samples\:within\:group}=\:n*\:\left((\sum\:_{k=1}^{s}k)\right.-\:\left.s\right)\:)\) and to all stimuli from all others groups ( \(\:{N}_{samples\:between\:group}=s*\left(s-1\right)*(\left(\sum\:_{k=1}^{n}k\right)-n)\) ) with n = number of groups and s = number of stimuli per group . Declarations Declaration of Interests The authors declare no competing interests. Funding This work was supported by Fonds Wetenschappelijk Onderzoek (FWO) grant G0B6422N and KU Leuven grant C14/22/134. T.T. is supported by FWO (senior clinical researcher; FWO 1830717N). M.V. holds an FWO fellowship for fundamental research (1169321N). Author Contribution M.V. conceived and designed the experiments. T.T. and M.V. planned and performed array implantations. M.V. performed the recordings. M.V. performed all clinical trial related activities. M.V. analyzed the data and wrote the manuscript. T.T. and P.J. supervised and guided the study. All authors reviewed and edited the manuscript. Acknowledgement We are grateful to the patient for her participation. We thank Stijn Verstraeten and Wouter Depuydt for technical assistance with the recording setup, Anaïs Van Hoylandt for assisting with data recording and data monitoring and Thomas Decramer for fine-tuning the implantation and recording technique from Utah-arrays in humans. We thank Rufin Vogels for providing the simple shape stimuli used in array 1. Data Availability The raw data is uploaded to a figshare repository at https://doi.org/10.6084/m9.figshare.30337921. References Ito, M., Tamura, H., Fujita, I. & Tanaka, K. Size and position invariance of neuronal responses in monkey inferotemporal cortex. J. Neurophysiol. 73 , 218–226. https://doi.org/10.1152/jn.1995.73.1.218 (1995). De Op, H. & Vogels, R. Spatial sensitivity of macaque inferior temporal neurons. J. Comp. Neurol. 426 , 505–518. https://doi.org/10.1002/1096-9861(20001030)426:4%3C505::aid-cne1%3E3.0.co;2-m (2000). Sereno, A. B. & Maunsell, J. H. Shape selectivity in primate lateral intraparietal cortex. Nature 395 , 500–503. https://doi.org/10.1038/26752 (1998). Sary, G., Vogels, R. & Orban, G. A. Cue-invariant shape selectivity of macaque inferior temporal neurons. Science 260 , 995–997. https://doi.org/10.1126/science.8493538 (1993). Logothetis, N. K., Pauls, J. & Poggio, T. Shape representation in the inferior temporal cortex of monkeys. Curr. Biol. 5 , 552–563. https://doi.org/10.1016/s0960-9822(95)00108-4 (1995). Janssen, P., Srivastava, S., Ombelet, S. & Orban, G. A. Coding of shape and position in macaque lateral intraparietal area. J. Neurosci. 28 , 6679–6690. https://doi.org/10.1523/JNEUROSCI.0499-08.2008 (2008). DiCarlo, J. J. & Maunsell, J. H. Anterior inferotemporal neurons of monkeys engaged in object recognition can be highly sensitive to object retinal position. J. Neurophysiol. 89 , 3264–3278. https://doi.org/10.1152/jn.00358.2002 (2003). Yamane, Y., Carlson, E. T., Bowman, K. C., Wang, Z. & Connor, C. E. A neural code for three-dimensional object shape in macaque inferotemporal cortex. Nat. Neurosci. 11 , 1352–1360. https://doi.org/10.1038/nn.2202 (2008). Verhoef, B. E., Vogels, R. & Janssen, P. Contribution of inferior temporal and posterior parietal activity to three-dimensional shape perception. Curr. Biol. 20 , 909–913. https://doi.org/10.1016/j.cub.2010.03.058 (2010). Verhoef, B. E., Vogels, R. & Janssen, P. Inferotemporal cortex subserves three-dimensional structure categorization. Neuron 73 , 171–182. https://doi.org/10.1016/j.neuron.2011.10.031 (2012). Janssen, P., Vogels, R., Liu, Y. & Orban, G. A. At least at the level of inferior temporal cortex, the stereo correspondence problem is solved. Neuron 37 , 693–701. https://doi.org/10.1016/s0896-6273(03)00023-0 (2003). Ramirez, J. G. et al. Intracortical recordings reveal the neuronal selectivity for bodies and body parts in the human visual cortex. Proc. Natl. Acad. Sci. U S A . 121 , e2408871121. https://doi.org/10.1073/pnas.2408871121 (2024). Decramer, T. et al. Single-cell selectivity and functional architecture of human lateral occipital complex. PLoS Biol. 17 , e3000280. https://doi.org/10.1371/journal.pbio.3000280 (2019). Decramer, T. et al. Single-Unit Recordings Reveal the Selectivity of a Human Face Area. J. Neurosci. 41 , 9340–9349. https://doi.org/10.1523/JNEUROSCI.0349-21.2021 (2021). Bougou, V. et al. Neuronal tuning and population representations of shape and category in human visual cortex. Nat. Commun. 15 , 4608. https://doi.org/10.1038/s41467-024-49078-3 (2024). Vogels, R. More Than the Face: Representations of Bodies in the Inferior Temporal Cortex. Annu. Rev. Vis. Sci. 8 , 383–405. https://doi.org/10.1146/annurev-vision-100720-113429 (2022). Afraz, S. R., Kiani, R. & Esteky, H. Microstimulation of inferotemporal cortex influences face categorization. Nature 442 , 692–695. https://doi.org/10.1038/nature04982 (2006). Malach, R. et al. Object-related activity revealed by functional magnetic resonance imaging in human occipital cortex. Proc. Natl. Acad. Sci. U S A . 92 , 8135–8139. https://doi.org/10.1073/pnas.92.18.8135 (1995). Kriegeskorte, N. et al. Matching categorical object representations in inferior temporal cortex of man and monkey. Neuron 60 , 1126–1141. https://doi.org/10.1016/j.neuron.2008.10.043 (2008). Grill-Spector, K. K. & Kanwisher, Z. The lateral occipital complex and its role in object recognition. Vis. Res. 41 , 1409–1422 (2001). Grill-Spector, K. et al. Differential processing of objects under various viewing conditions in the human lateral occipital complex. Neuron 24 , 187–203. https://doi.org/10.1016/s0896-6273(00)80832-6 (1999). Sayres, R. & Grill-Spector, K. Relating retinotopic and object-selective responses in human lateral occipital cortex. J. Neurophysiol. 100 , 249–267. https://doi.org/10.1152/jn.01383.2007 (2008). Rainer, G., Asaad, W. F. & Miller, E. K. Selective representation of relevant information by neurons in the primate prefrontal cortex. Nature 393 , 577–579. https://doi.org/10.1038/31235 (1998). Cichy, R. M., Chen, Y. & Haynes, J. D. Encoding the identity and location of objects in human LOC. Neuroimage 54 , 2297–2307. https://doi.org/10.1016/j.neuroimage.2010.09.044 (2011). Kravitz, D. J., Kriegeskorte, N. & Baker, C. I. High-level visual object representations are constrained by position. Cereb. Cortex . 20 , 2916–2925. https://doi.org/10.1093/cercor/bhq042 (2010). Hemond, C. C. & Kanwisher, N. G. Op de Beeck, H. P. A preference for contralateral stimuli in human object- and face-selective cortex. PLoS One . 2 , e574. https://doi.org/10.1371/journal.pone.0000574 (2007). MacEvoy, S. P. & Epstein, R. A. Position selectivity in scene- and object-responsive occipitotemporal regions. J. Neurophysiol. 98 , 2089–2098. https://doi.org/10.1152/jn.00438.2007 (2007). Cichy, R. M. et al. Probing principles of large-scale object representation: category preference and location encoding. Hum. Brain Mapp. 34 , 1636–1651. https://doi.org/10.1002/hbm.22020 (2013). Downing, P. E., Jiang, Y., Shuman, M. & Kanwisher, N. A cortical area selective for visual processing of the human body. Science 293 , 2470–2473. https://doi.org/10.1126/science.1063414 (2001). Popivanov, I. D., Jastorff, J., Vanduffel, W. & Vogels, R. Tolerance of macaque middle STS body patch neurons to shape-preserving stimulus transformations. J. Cogn. Neurosci. 27 , 1001–1016. https://doi.org/10.1162/jocn_a_00762 (2015). Hung, C. P., Kreiman, G., Poggio, T. & DiCarlo, J. J. Fast readout of object identity from macaque inferior temporal cortex. Science 310 , 863–866. https://doi.org/10.1126/science.1117593 (2005). Rust, N. C. & Dicarlo, J. J. Selectivity and tolerance (invariance) both increase as visual information propagates from cortical area V4 to IT. J. Neurosci. 30 , 12978–12995. https://doi.org/10.1523/JNEUROSCI.0179-10.2010 (2010). Tootell, R. B., Mendola, J. D., Hadjikhani, N. K., Liu, A. K. & Dale, A. M. The representation of the ipsilateral visual field in human cerebral cortex. Proc. Natl. Acad. Sci. U S A . 95 , 818–824. https://doi.org/10.1073/pnas.95.3.818 (1998). Grill-Spector, K.et al. A sequence of object-processing stages revealed by fMRI in the human occipital lobe. Hum Brain Mapp 6, 316–328 (1998). https://doi.org/10.1002/(SICI)1097-0193(1998)6:43.0.CO;2-6 Schwarzlose, R. F., Swisher, J. D., Dang, S. & Kanwisher, N. The distribution of category and location information across object-selective regions in human visual cortex. Proc. Natl. Acad. Sci. U S A . 105 , 4447–4452. https://doi.org/10.1073/pnas.0800431105 (2008). Williams, M. A. et al. Feedback of visual object information to foveal retinotopic cortex. Nat. Neurosci. 11 , 1439–1445. https://doi.org/10.1038/nn.2218 (2008). Graumann, M., Ciuffi, C., Dwivedi, K., Roig, G. & Cichy, R. M. The spatiotemporal neural dynamics of object location representations in the human brain. Nat. Hum. Behav. 6 , 796–811. https://doi.org/10.1038/s41562-022-01302-0 (2022). Desimone, R., Albright, T. D., Gross, C. G. & Bruce, C. Stimulus-selective properties of inferior temporal neurons in the macaque. J. Neurosci. 4 , 2051–2062. https://doi.org/10.1523/JNEUROSCI.04-08-02051.1984 (1984). Kobatake, E. & Tanaka, K. Neuronal selectivities to complex object features in the ventral visual pathway of the macaque cerebral cortex. J. Neurophysiol. 71 , 856–867. https://doi.org/10.1152/jn.1994.71.3.856 (1994). Richmond, B. J., Wurtz, R. H. & Sato, T. Visual responses of inferior temporal neurons in awake rhesus monkey. J. Neurophysiol. 50 , 1415–1432. https://doi.org/10.1152/jn.1983.50.6.1415 (1983). Dale, A. M., Fischl, B. & Sereno, M. I. Cortical surface-based analysis. I. Segmentation and surface reconstruction. Neuroimage 9 , 179–194. https://doi.org/10.1006/nimg.1998.0395 (1999). Blenkmann, A. O. et al. iElectrodes: A Comprehensive Open-Source Toolbox for Depth and Subdural Grid Electrode Localization. Front. Neuroinform . 11 , 14. https://doi.org/10.3389/fninf.2017.00014 (2017). Popivanov, I. D., Jastorff, J., Vanduffel, W. & Vogels, R. Stimulus representations in body-selective regions of the macaque cortex assessed with event-related fMRI. Neuroimage 63 , 723–741. https://doi.org/10.1016/j.neuroimage.2012.07.013 (2012). Peelen, M. V. & Downing, P. E. Selectivity for the human body in the fusiform gyrus. J. Neurophysiol. 93 , 603–608. https://doi.org/10.1152/jn.00513.2004 (2005). Vanhoyland, M., Janssen, P. & Theys, T. Single-neuron activity reflects visual awareness in human lateral occipital complex. Preprint Res. Square . https://doi.org/10.21203/rs.3.rs-5941203/v1 (2025). https://doi.org/. Vanhoyland, M., Janssen, P. & Theys, T. Data from Human lateral occipital complex is invariant for position and size transformations at the single-neuron level figshare (2025). https://doi.org/10.6084/m9.figshare.30337921 Additional Declarations No competing interests reported. Supplementary Files SupplementaryfiguresHumanlateraloccipitalcomplexisinvariantforpositionandsizetransformationsatthesingleneuronlevel.docx Cite Share Download PDF Status: Published Journal Publication published 12 Mar, 2026 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Revision requested 28 Jan, 2026 Reviews received at journal 28 Jan, 2026 Reviewers agreed at journal 12 Jan, 2026 Reviews received at journal 23 Dec, 2025 Reviewers agreed at journal 03 Dec, 2025 Reviewers invited by journal 03 Dec, 2025 Editor assigned by journal 03 Dec, 2025 Editor invited by journal 02 Dec, 2025 Submission checks completed at journal 01 Dec, 2025 First submitted to journal 01 Dec, 2025 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-8222241","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":555392979,"identity":"772d9998-7a57-4b10-8789-08f6dca88ad5","order_by":0,"name":"Vanhoyland Michaël","email":"data:image/png;base64,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","orcid":"","institution":"KU Leuven","correspondingAuthor":true,"prefix":"","firstName":"Vanhoyland","middleName":"","lastName":"Michaël","suffix":""},{"id":555392980,"identity":"02806e74-ef06-4c2a-adf0-995c155202b1","order_by":1,"name":"Janssen Peter","email":"","orcid":"","institution":"Leuven Brain Institute","correspondingAuthor":false,"prefix":"","firstName":"Janssen","middleName":"","lastName":"Peter","suffix":""},{"id":555392981,"identity":"8312d6ae-8439-4085-b19e-7e1356548362","order_by":2,"name":"Theys Tom","email":"","orcid":"","institution":"Universitair Ziekenhuis Leuven","correspondingAuthor":false,"prefix":"","firstName":"Theys","middleName":"","lastName":"Tom","suffix":""}],"badges":[],"createdAt":"2025-11-27 12:53:21","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8222241/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8222241/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-026-43946-2","type":"published","date":"2026-03-12T16:00:21+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":97674588,"identity":"db4ebe10-107a-4099-844e-abf046b0ceec","added_by":"auto","created_at":"2025-12-08 09:43:38","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2830498,"visible":true,"origin":"","legend":"","description":"","filename":"ManuscriptHumanlateraloccipitalcomplexisinvariantforpositionandsizetransformationsatthesingleneuronlevelfiguresembedded.docx","url":"https://assets-eu.researchsquare.com/files/rs-8222241/v1/654a0f8608bef51cb95a8db8.docx"},{"id":97674615,"identity":"3f0e6e3a-e685-465d-a81e-e18997d86ae5","added_by":"auto","created_at":"2025-12-08 09:43:43","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":4935,"visible":true,"origin":"","legend":"","description":"","filename":"89558d142484456cbdbf0fe1867a7614.json","url":"https://assets-eu.researchsquare.com/files/rs-8222241/v1/047a6d8275f16a4bf06969cb.json"},{"id":97674562,"identity":"1f69dd66-0fcb-4c08-8b75-3ea0748e75ff","added_by":"auto","created_at":"2025-12-08 09:43:37","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1892400,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryfiguresHumanlateraloccipitalcomplexisinvariantforpositionandsizetransformationsatthesingleneuronlevel.docx","url":"https://assets-eu.researchsquare.com/files/rs-8222241/v1/c310997e03c69be736755eed.docx"},{"id":97658251,"identity":"2c12232b-7aa6-4c9f-b1a3-8abdaa2b5eda","added_by":"auto","created_at":"2025-12-08 07:31:49","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":141495,"visible":true,"origin":"","legend":"","description":"","filename":"89558d142484456cbdbf0fe1867a76141enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-8222241/v1/6104c7177717897fccc84202.xml"},{"id":97658260,"identity":"ec225c08-f72c-4645-8a73-2732e852adc0","added_by":"auto","created_at":"2025-12-08 07:31:49","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":62900,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8222241/v1/b3011351eff44f09ffb44e47.png"},{"id":97658249,"identity":"842be1ec-6525-4199-8ffa-5dbfaa89189e","added_by":"auto","created_at":"2025-12-08 07:31:48","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":68687,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8222241/v1/399cbbc5bbb69dd7f785d2dd.png"},{"id":97673222,"identity":"02e5f238-c78a-4e45-ad56-2e8ad4bf2976","added_by":"auto","created_at":"2025-12-08 09:39:41","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":36689,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8222241/v1/e46d79b994ec0542b5eb32a4.png"},{"id":97658263,"identity":"71b8626e-fd9e-4758-8148-288c777e9aff","added_by":"auto","created_at":"2025-12-08 07:31:49","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":94642,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8222241/v1/946d00d985d84fcc51f91536.png"},{"id":97672501,"identity":"826ca32d-833c-475b-be89-1b9550426551","added_by":"auto","created_at":"2025-12-08 09:38:10","extension":"png","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":54406,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-8222241/v1/61299772ec64074712da98aa.png"},{"id":97658261,"identity":"4d21f9be-0d5a-4dde-b681-dc9cd9b6c580","added_by":"auto","created_at":"2025-12-08 07:31:49","extension":"png","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":173045,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-8222241/v1/d392f37f0f7ce30c5ee8f540.png"},{"id":97658259,"identity":"dced4388-1914-4dad-8b7e-2cf8fd811ae3","added_by":"auto","created_at":"2025-12-08 07:31:49","extension":"png","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":25178,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-8222241/v1/af8c776532604f8994dfc570.png"},{"id":97658264,"identity":"86b9f19b-c0a9-4545-b26d-ebd97c243c57","added_by":"auto","created_at":"2025-12-08 07:31:49","extension":"xml","order_by":18,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":138761,"visible":true,"origin":"","legend":"","description":"","filename":"89558d142484456cbdbf0fe1867a76141structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-8222241/v1/b1944f11bace3a708228c042.xml"},{"id":97658262,"identity":"25f3bcb4-02fd-4634-82db-24c4fbd9cc28","added_by":"auto","created_at":"2025-12-08 07:31:49","extension":"html","order_by":19,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":155384,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-8222241/v1/b4c0f2b20e4585835021de1e.html"},{"id":97658244,"identity":"62d3837c-726d-4499-b7e5-b12d10156cc0","added_by":"auto","created_at":"2025-12-08 07:31:48","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":211103,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMulti-electrode array recording sites and experimental paradigm for determination of invariance within receptive field. \u003c/strong\u003e\u0026nbsp;\u003cstrong\u003eA)\u003c/strong\u003e Anatomical locations of Utah arrays projected on 3D cortical renderings (left hemisphere). MNI coordinates (X, Y, Z) were as follows: A1 (-41, -83, 9) and A2 (-38, -84, -5). Orange dotted lines represent lateral occipital sulcus. Arrow points to pre-occipital notch. \u003cstrong\u003eB)\u003c/strong\u003e Positions\u003cstrong\u003e \u003c/strong\u003eon the screen of stimulus presentation when testing for positional invariance in both arrays. Below each monitor is a subset of the presented stimuli shown (8 out of 64 for array 1, all for array 2). All 64 shapes are presented in Figure S1. \u003cstrong\u003eC)\u003c/strong\u003ePositions\u003cstrong\u003e \u003c/strong\u003eof stimulus presentation when testing for size invariance in array 2. The 3 contours represent 2°, 4° and 8°. Ten of 20 torsos that were used are shown below the monitor. Remaining 10 torsos are presented below B.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8222241/v1/670194a615d917bf123279a4.png"},{"id":97658246,"identity":"50f0deb8-f70d-421c-9f34-83a09d24f1d2","added_by":"auto","created_at":"2025-12-08 07:31:48","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":477618,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eReceptive field. A) \u003c/strong\u003eAverage receptive field for both arrays. \u003cstrong\u003eB)\u003c/strong\u003eBubble plot showing the center of the receptive field of each single unit (blue, filled circles) and other locations that are part of the receptive field (red, empty circles; response ≥50% of center). Exact location of the dots within each 4° by 4° square was randomized. \u003cstrong\u003eC)\u003c/strong\u003e Example receptive fields of 4 single units (two most left from array 1, two most right from array 2).\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8222241/v1/b5c4e70fa4e43da3a514468b.png"},{"id":97672469,"identity":"aaafacee-d297-4970-a381-ad56211d7450","added_by":"auto","created_at":"2025-12-08 09:38:04","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":329551,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eShape selectivity profile. A) \u003c/strong\u003eAverage response (\u003cimg width=\"11\" height=\"17\" src=\"data:image/png;base64,R0lGODlhEAAaAHcAMSH+GlNvZnR3YXJlOiBNaWNyb3NvZnQgT2ZmaWNlACH5BAEAAAAALAEABwAPAA8AggAAAAAAAAAAZgA6kLb//9uQOtv///+2ZgMlCLoltDC2F+us1mGpN+vQEYxkOQ5GRnnAxbag57KzHL+VaO5oAgA7\"/\u003e\u0026nbsp;standard error) of all responsive single units (N\u003csub\u003eA1 \u003c/sub\u003e= 22; N\u003csub\u003eA2 \u003c/sub\u003e= 26) to the most and least preferred stimulus\u003cstrong\u003e \u003c/strong\u003eat the preferred location. Sorting between stimuli was done on the 150 ms response window. Vertical dotted line represents stimulus onset. Horizontal bars indicate significance between both graphs (p \u0026lt;0.05, two-sided independent samples t-test with at least three consecutive significant time points)\u0026nbsp;\u0026nbsp;\u003cstrong\u003eB)\u003c/strong\u003e Histograms displaying the distribution of the shape tuning index (left) and selectivity depth (right) of all responsive single units in both arrays.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8222241/v1/85dc81e1b4800603718d4096.png"},{"id":97658255,"identity":"8eaf80fb-db25-4adf-9a1a-d0069d1d64e3","added_by":"auto","created_at":"2025-12-08 07:31:49","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":927759,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eExample single unit with high shape selectivity and strong response invariance within receptive field. A)\u003c/strong\u003e Net spike rate within response window (200-250 ms) to all 10 torsos presented in all 5 positions. Torsos in each position are sorted according to the sorting order obtained in the preferred position. \u003cstrong\u003eB) \u003c/strong\u003eWaveform. \u003cstrong\u003eC) \u003c/strong\u003eAverage response within the response window to 5 torsos within the 5 tested positions. Torsos are presented below D.\u003cstrong\u003e D) \u003c/strong\u003eRaster plots of 5 torsos presented in 5 positions. Each line on the y-axis represents a different trial. Raster plots are color coded with the net spike rate, smoothed with a moving average of 100 ms.\u003cstrong\u003e \u003c/strong\u003eVertical lines represent stimulus onset.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8222241/v1/68105a68f041cbf027644009.png"},{"id":97674590,"identity":"2a9255d4-371d-4f24-8a4e-01293a934098","added_by":"auto","created_at":"2025-12-08 09:43:38","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":353320,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eInvariance for position within the receptive field. A)\u003c/strong\u003e Time-resolved Pearson correlation between sorted stimuli in the preferred and non-preferred positions (inter position: green line). The blue line represents intertrial variability, providing the theoretical maximum possible correlation for the green line. Shaded regions indicate 95% confidence intervals. Horizontal bars mark time intervals where the inter-position correlation was significantly higher than expected under random label shuffling (p \u0026lt; 0.05, one-sided permutation test with a minimum of three consecutive significant time points). \u003cstrong\u003eB)\u003c/strong\u003e Scatter plot illustrating the responses in the preferred location (x-axis) against the corresponding sorted responses in non-preferred locations (y-axis) at peak correlation (Array 1: 210–310 ms; Array 2: 180–280 ms). Dotted line represents a fitted linear regression. \u003cstrong\u003eC)\u003c/strong\u003e Average sorted Z-normalized responses across all single units in the preferred location (orange), alongside the corresponding average Ordinary Least Squares (OLS) linear regression for the non-preferred locations (blue). \u003cstrong\u003eD) \u003c/strong\u003eTwo-example units from array 1 where shapes in the non-preferred position (blue) are sorted according to the sorting order obtained from the preferred position (orange). Blue line is the OLS linear regression.\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-8222241/v1/3311a949ad779e4d2e805f7c.png"},{"id":97674566,"identity":"b77121da-af30-4e32-82f2-3cbeb0da9dee","added_by":"auto","created_at":"2025-12-08 09:43:37","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":677423,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMultidimensional scaling. A) \u003c/strong\u003eVisualization of similarity relations 210-310 ms after stimulus onset in array 1 (left) \u0026nbsp;and 180 - 280 ms after stimulus onset in array 2 (right) (time corresponds to peak of correlation, see Figure 5A). Red border: color coding according to position on screen. Blue border: color coding according to individual stimulus. Pearson’s r measures the correlation between the multi-dimensional space and two-dimension embedding. \u003cstrong\u003eB) \u003c/strong\u003eAverage Euclidean distance over time \u003cem\u003ewithin \u003c/em\u003egroups (yellow) versus \u003cem\u003ebetween \u003c/em\u003egroups (blue) as determined by multidimension scaling. Shading represents 95% confidence intervals. Vertical dotted line represents stimulus onset. The x-axis represents the middle of the 100 ms bin used for the multidimensional scaling. As visualized at the right side, where the circles represent different positions of stimuli presentation, the upper plots compare the distance \u003cem\u003ewithin\u003c/em\u003ethe stimulus presentation position (yellow) to \u003cem\u003ebetween\u003c/em\u003ethe other locations (blue), whereas the lower plots compare the distance of the same shape presented at different positions (yellow) to the other stimuli presented at the other positions (blue). Horizontal bars indicate significance between yellow and blue graphs (p \u0026lt;0.05, two-sided independent samples t-test with at least five consecutive significant time points). Both position (upper plots) and shape identity (lower plots) have a significant effect on the multi-dimensional scaling. The number of pairwise distances of which the average is taken, is presented in each plot in the upper left corner.\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-8222241/v1/f298e8fb5d776ff8569746da.png"},{"id":97658257,"identity":"a3c233b0-ec60-46b1-984f-82219ba364a7","added_by":"auto","created_at":"2025-12-08 07:31:49","extension":"jpeg","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":139716,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eInvariance for changes in size. A) \u003c/strong\u003eExample single unit responsive for all 3 sizes. Vertical lines separate different sizes. Torsos in all sizes are sorted according to the sorting order obtained from 8°\u003csup\u003e \u003c/sup\u003e(size eliciting the strongest response).\u003cstrong\u003e B) \u003c/strong\u003eTime-resolved Pearson correlation between sorted stimuli in the size eliciting the strongest response and the other sizes (inter size: green line). Different size comparisons are pooled (8° vs 4°,8° vs 2°, 4° vs 2°). Other conventions are same as in figure 5A. \u003cstrong\u003eC) \u003c/strong\u003eTime-resolved Pearson correlation between ranked stimuli in separate sizes for neurons that were responsive to \u003cimg width=\"11\" height=\"17\" src=\"data:image/png;base64,R0lGODlhEAAaAHcAMSH+GlNvZnR3YXJlOiBNaWNyb3NvZnQgT2ZmaWNlACH5BAEAAAAALAEACAAPAA0AhAAAAAAAAAAAOgA6OgA6ZgA6kDoAADo6ZjpmkDpmtmY6AGaQtma225BmOpC227aQZrbb29uQZtu2kNvbttvb/9v///+2Zv+2kP/b2///2wECAwECAwECAwECAwECAwECAwU94LVUQGmeZfYYB4S+5dQICQWjmaQMDHmbmBXC9TNFDAXfK2gYwnIKgkN5ktFsLxWLeLv0iuCXJUAum8nJEAA7\"/\u003e\u0026nbsp;1 torso in both sizes. Horizontal bars mark time intervals where the correlation was significantly higher than expected under random label shuffling (p \u0026lt; 0.05, one-sided permutation test). \u003cstrong\u003eD)\u003c/strong\u003e Average sorted Z-normalized responses across all single units in the size eliciting the strongest response (orange), alongside the corresponding average Ordinary Least Squares (OLS) linear regression for the other responsive sizes (blue).\u003c/p\u003e","description":"","filename":"floatimage7.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8222241/v1/c4cdc6a2cb6109e84aa3eb91.jpeg"},{"id":104740132,"identity":"f06907b5-bedd-44a6-a802-a8f54bd8b028","added_by":"auto","created_at":"2026-03-16 16:15:28","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":4003753,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8222241/v1/5141ed6d-ac17-4ca1-86a2-ce60d74aa1d1.pdf"},{"id":97658247,"identity":"d9193fda-655f-4582-9619-c4f984faa7b0","added_by":"auto","created_at":"2025-12-08 07:31:48","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":1892400,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryfiguresHumanlateraloccipitalcomplexisinvariantforpositionandsizetransformationsatthesingleneuronlevel.docx","url":"https://assets-eu.researchsquare.com/files/rs-8222241/v1/eed011b5f19397c00e0e2e77.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Human lateral occipital complex is invariant for position and size transformations at the single-neuron level","fulltext":[{"header":"Significance statement","content":"\u003cp\u003eMacaques have served as a valuable animal model to better understand (human) brain functions across various regions, including shape selectivity and tolerance to object manipulations. However, the extent to which these findings apply to humans remains unclear. While non-invasive recording techniques can address some questions, studying shape selectivity requires measuring responses from individual neurons, which necessitates invasive intracortical recordings. Traditional depth electrodes, which target medial brain structures, are ill-suited for recording single neurons in the lateral occipital complex (LO), which is located on the brain's convexity. To overcome this, we implanted two micro-electrode arrays in different LO subregions and demonstrated that individual neurons exhibit tolerance to changes in position and size, mirroring patterns observed in the macaque inferotemporal cortex.\u003c/p\u003e"},{"header":"Introduction","content":"\u003cp\u003eA critical aspect of object recognition is invariance to stimulus transformations such as changes in location and size. For instance, consider a table with chairs around it: with a single glance, you can effortlessly categorize all chairs as \"chairs\" despite their varying positions relative to your viewpoint.\u003c/p\u003e\u003cp\u003eNeural mechanisms underlying invariances have been extensively studied in monkeys across the ventral (notably areas TE and TEO in the inferotemporal cortex (ITC)) and the dorsal visual stream (lateral intraparietal cortex, LIP).\u003csup\u003e\u003cspan additionalcitationids=\"CR2 CR3 CR4 CR5 CR6 CR7\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e–\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e Macaque ITC, which shares many similarities with the human lateral occipital complex (LO) with regard to shape processing,\u003csup\u003e9–19\u003c/sup\u003e maintains stimulus selectivity across the receptive field and tolerates size modulations at the single-neuron level.\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e,\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e Extrapolating these findings to humans has been hindered by the need for single-neuron recordings to accurately investigate invariances. Activity in human LO, a pivotal region in the ventral stream for shape and object recognition,\u003csup\u003e18,20\u003c/sup\u003e undergoes adaptation in fMRI when a stimulus is repeatedly shown while simultaneously being changed in size or position, suggesting response invariance at the population level.\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e LO displays strong shape selectivity at the single-neuron level and features large receptive fields that can span contra-, ipsi-, or bilateral regions.\u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e,\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e Within LO, patches of face- and body-selective cells have been identified,\u003csup\u003e12–14\u003c/sup\u003e though category responses appear primarily driven by underlying shape features.\u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e,\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e Recently, we investigated two such invariances in LO and found that body-selectivity generalized for silhouettes and showed partial tolerance for rotational manipulations.\u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e Other invariances, however, have thus far not been investigated.\u003c/p\u003e\u003cp\u003eTo address this, we recorded single-unit activity from two microelectrode arrays implanted in distinct LO subregions in the same patient. Our findings confirm the heterogeneous nature of LO receptive fields while revealing that these fields are smaller than previously reported.\u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e Employing time-resolved correlational analyses and multidimensional scaling (MDS), we demonstrate that stimulus selectivity of a population of LO neurons is invariant to positional shifts within receptive fields and show partial tolerance to size manipulations.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003eReceptive field mapping\u003c/h2\u003e\u003cp\u003eWe recorded 28 visually responsive neurons using 2 multi-electrode arrays (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{N}_{A1}=13\\:;{N}_{A2}=\\:15)\\)\u003c/span\u003e\u003c/span\u003e implanted in shape-selective lateral occipital cortex (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Sensitivity to image scrambling in both arrays was demonstrated in previous research, thereby confirming that both arrays were located in LO.\u003csup\u003e15\u003c/sup\u003e Moreover, array 2 was implanted in a body patch.\u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e We mapped the central 28\u0026deg; \u0026times; 28\u0026deg; receptive field by presenting a headless torso at 49 non-overlapping positions. For A1, the average receptive field was (para)foveal along the vertical meridian, while for A2, it was highly eccentric in the contralateral (right), inferior quadrant (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA). An overview of the number of neurons responding in each quadrant is presented in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. In A1, 67% of neurons responded foveally, compared to only 1 neuron (8%) in A2 (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB). Both arrays showed a contralateral bias, with 60% of neurons in A1 and 92% in A2 responding contralaterally. The average receptive field per array, including the positions of the receptive field centers per single-unit, are presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA and \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB. The average size of the receptive field was 70 deg\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e and 98 deg\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e in A1 and A2, respectively.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\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\u003eNumber of single-units of which the receptive field was mapped, and the position of the receptive field.\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=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colspan=\"2\" nameend=\"c4\" namest=\"c3\"\u003e\u003cp\u003eArray 1\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"2\" nameend=\"c6\" namest=\"c5\"\u003e\u003cp\u003eArray 2\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003ctr\u003e\u003cth align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eN\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003e%\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eN\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003e%\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e\u003cp\u003eResponsive\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e15\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e100\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e13\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e100\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"2\" nameend=\"c2\" namest=\"c1\"\u003e\u003cp\u003eFoveal\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e67\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e8\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e\u003cp\u003eLaterality\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eIpsilateral\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e20\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eContralateral\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e60\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e92\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eBilateral\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e8\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\" morerows=\"2\" rowspan=\"3\"\u003e\u003cp\u003eVerticality\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eUpper hemifield\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e20\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e8\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eLower hemifield\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e27\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e46\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eBoth\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e53\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e38\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eShape selectivity\u003c/h3\u003e\n\u003cp\u003eGiven the differences in stimulus selectivity and receptive fields across the two arrays, we presented distinct stimulus sets at different positions within the average receptive field, as displayed in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB. For A1, we presented 64 simple shapes (presented in Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e) at two positions along the vertical meridian (above and below the fixation marker). For array 2, we displayed 10 torso images at five locations in the contralateral lower quadrant. To ensure stable fixation, trials were automatically aborted if the gaze deviated beyond a predefined 3\u0026deg; \u0026times; 3\u0026deg; fixation window. Furthermore, we verified that the average change in eye position during stimulus presentation (from onset to offset) was similar for different positions (Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e-2). This verification was particularly critical for array 1, where stimuli were presented closer to the fixation marker, as saccades toward the stimulus locations could compromise data interpretation.\u003c/p\u003e\u003cp\u003eWe recorded 48 neurons that were responsive in at least 2 locations (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{N}_{A1}=22\\:;{N}_{A2}=26).\\)\u003c/span\u003e\u003c/span\u003e For each neuron, we determined the preferred stimulus location (see Methods for criteria) and sorted the different stimuli according to the responses at the preferred location. As a measure of shape selectivity, we calculated the shape tuning index and depth of selectivity (see Methods).\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e,\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e In short, the shape tuning index (STI) measures how broadly a neuronal response is tuned for shapes by taking the net response to the best and worst stimulus into account, where STI\u0026thinsp;=\u0026thinsp;1 signifies very strong tuning, and STI\u0026thinsp;=\u0026thinsp;0 indicates no selectivity. In contrast, the selectivity depth takes all responses to all stimuli into account and is 1 if a neuron responds to only 1 stimulus, and 0 if it responds to all stimuli equally. The average STI in arrays 1 and 2 was 1.15 \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\pm\\:\\)\u003c/span\u003e\u003c/span\u003e 0.03 and 0.93 \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\pm\\:\\)\u003c/span\u003e\u003c/span\u003e 0.03, respectively. (Values \u0026gt;1 are caused by inhibitory responses to the least preferred stimulus.) The mean selectivity depth measured 0.79 \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\pm\\:\\)\u003c/span\u003e\u003c/span\u003e 0.02 and 0.67 \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\pm\\:\\)\u003c/span\u003e\u003c/span\u003e 0.03, respectively. The distribution of STI and selectivity depth are presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB. Both indices point towards strong shape tuning, which is further visualized in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA where the average response to the worst stimulus is negligible, while the preferred stimulus evoked on average a net increase of 25\u0026ndash;40 spikes/s.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\n\u003ch3\u003ePosition invariance of the selectivity within the receptive field\u003c/h3\u003e\n\u003cp\u003eTo investigate whether the shape preference is maintained within the receptive field, the stimuli at the other, non-preferred location(s) were ranked according to the preference at the preferred location. An example single unit with strong shape selectivity (STI\u0026thinsp;=\u0026thinsp;1.05; selectivity depth\u0026thinsp;=\u0026thinsp;0.87) and strong selectivity invariance across the receptive field is presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e. For all shape-selective neurons (N\u0026thinsp;=\u0026thinsp;48), we calculated the Pearson correlation coefficient \u0026lsquo;r\u0026rsquo; between the ranked responses at the preferred and non-preferred locations over time after stimulus onset. In case of selectivity invariance at the population level, a significant correlation exists between preferred and non-preferred locations. For both arrays, the ranked responses in the preferred and non-preferred locations were strongly correlated starting from 90 ms and 80 ms (middle of 100 ms bin) in array 1 and 2, respectively (peak correlation in array 1: r\u0026thinsp;=\u0026thinsp;0.32, p\u0026thinsp;=\u0026thinsp;3.62 x 10\u003csup\u003e\u0026ndash;34\u003c/sup\u003e; array 2: r\u0026thinsp;=\u0026thinsp;0.57, p\u0026thinsp;=\u0026thinsp;1.07 x 10\u003csup\u003e\u0026ndash;83\u003c/sup\u003e, two-sided t-test, Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA). The lower correlations obtained in array 1 are partially due to the larger number of used shapes. We therefore repeated the analysis with 10 stimuli (5 best and 5 worst per neuron) to match the number of stimuli in array 2, which resulted in correlations similar to array 2 (r\u0026thinsp;=\u0026thinsp;0.53, 2.95 x 10\u003csup\u003e\u0026ndash;19\u003c/sup\u003e, two-sided t-test, Figure S3). Because responses are variable between trials of the same stimulus, we also calculated the intertrial correlations (see Methods), which serves as theoretical maximal obtainable correlation per array. For both arrays, the intertrial correlations exceeded the correlations between preferred and non-preferred location (peak correlation in array 1: r\u0026thinsp;=\u0026thinsp;0.52 p\u0026thinsp;=\u0026thinsp;5.54 x 10\u003csup\u003e\u0026ndash;94\u003c/sup\u003e; array 2: r\u0026thinsp;=\u0026thinsp;0.73; p\u0026thinsp;=\u0026thinsp;4.69 x 10\u003csup\u003e\u0026ndash;57\u003c/sup\u003e, two-sided t-test, Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA).\u003c/p\u003e\u003cp\u003eFor each neuron, we fitted the ordinary least squares (OLS) linear regression through the ranked response in the non-preferred locations (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC). In case of selectivity invariance within the receptive field, the slope of the regression should be significantly different from 0 (e.g. negative slope in case of descending ranking). For both arrays, the average slope was strictly negative (Array 1: \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\beta\\:\\)\u003c/span\u003e\u003c/span\u003e = -0.454 \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\pm\\:\\:0.075\\)\u003c/span\u003e\u003c/span\u003e, Array 2: \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\beta\\:\\)\u003c/span\u003e\u003c/span\u003e = -0.627 \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\pm\\:\\)\u003c/span\u003e\u003c/span\u003e 0.050). Lastly, we performed multidimensional scaling to visualize, in an unbiased way, if a stimulus would be grouped together despite being presented in different locations on the screen (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e). The position in which a stimulus was presented on the screen largely influenced the position on the two-dimensional representation, which is a direct result of receptive field interactions. However, the distance in the two-dimensional space was smaller between identical shapes presented at different positions compared to different shapes presented at different positions, signifying selectivity invariance (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB).\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\n\u003ch3\u003eSize invariance\u003c/h3\u003e\n\u003cp\u003eWe presented 20 torsos in 3 different sizes (2, 4 and 8 visual degrees) in the receptive field (contralateral lower quadrant) and recorded 29 visually responsive single units on array 2, of which 14 were responsive to \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\ge\\:\\)\u003c/span\u003e\u003c/span\u003e 1 size (14 neurons were only responsive to the 8\u0026deg; torsos). Ten neurons were only responsive to 4\u0026deg; and 8\u0026deg;, 1 neuron to only 2\u0026deg; and 8\u0026deg; and 3 neurons to all 3 sizes. The responses from a neuron responsive to all 3 sizes is presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eA. Similar to the analysis done for position invariance, we ranked the responses to different sized torso\u0026rsquo;s according to sorting order obtained from the size eliciting the strongest response. Subsequently, we calculated the time-resolved Pearson correlation (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eB-C) and the slope from the ordinary least squares linear regression (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eD). Peak correlation was obtained in the 270\u0026ndash;370 ms interval (r = 0.36, p = 1.28 x 10\u003csup\u003e\u0026ndash;11\u003c/sup\u003e, two-sided t-test). Because this analysis pooled all size pairs (2\u0026deg; vs 4\u0026deg;, 2\u0026deg; vs 8\u0026deg; and 4\u0026deg; vs 8\u0026deg;), we also calculated the correlation between each pair individually to exclude that the main effect was carried by the correlations from only 1 pair. As visualized in Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eC, we obtained significant correlations between the sorted responses for all pairs (2\u0026deg; vs 4\u0026deg;: r = 0.29, p = 1.13 x 10\u003csup\u003e\u0026minus;\u0026thinsp;9\u003c/sup\u003e, 2\u0026deg; vs 8\u0026deg;: r = 0.28, p = 2.12 x 10\u003csup\u003e\u0026minus;\u0026thinsp;5\u003c/sup\u003e; 4\u0026deg; vs 8\u0026deg;: r = 0.40, p =\u0026thinsp;3.84 x 10\u003csup\u003e\u0026ndash;11\u003c/sup\u003e; two-sided t-test). Finally, also the slope from the OLS linear regression was significantly negative (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\beta\\:\\)\u003c/span\u003e\u003c/span\u003e = -0.441 \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\pm\\:\\)\u003c/span\u003e\u003c/span\u003e 0.076), indicating size invariance at the population level.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eWe conducted the first single-unit recordings in the human lateral occipital complex (LO) to investigate neuronal response invariance to positional shifts and size variations within the receptive field. Our findings demonstrate that LO neurons exhibit strong shapes tuning and robust tolerance to positional changes. Shape selectivity was preserved across size variations of up to two octaves (fourfold size increase). The invariance of the shape selectivity across stimulus transformations highlights LO’s crucial role in object recognition.\u003c/p\u003e\u003cp\u003eDue to the semi-acute nature of our multielectrode recordings in a human subject, mapping the precise receptive field for each neuron individually with its optimal stimulus was not feasible. Instead, we estimated the average receptive field, tested various shapes at different positions within this field, and presented shapes of varying sizes at the center of the average receptive field to maximize the number of responsive units. Non-responsive positions and sizes were excluded from pairwise analysis, as they were outside the receptive field. Notably, nearly all neurons (28 of 29) responded to 8° × 8° shapes, while only 4 of 29 responded to 2° × 2° shapes. This discrepancy may suggest suboptimal centering within the receptive field, or that the smaller shapes were too small to reliably evoke responses. Additionally, the 8° × 8° shapes may have extended beyond some receptive field boundaries. Estimated average receptive field sizes were 70 deg² and 98 deg² for array 1 and 2, respectively. However, since some neurons in array 2 responded to the most eccentric positions as visualized in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC (+ 14° on the x-axis, and − 14° on the y-axis), the true receptive field size for array 2 might be underestimated. Our estimates are notably smaller than a prior study’s reported 473 deg².\u003csup\u003e13\u003c/sup\u003e This could possibly be explained by differences in used stimuli (4° × 4° torsos here vs. 10° × 6° shapes in Decramer \u0026amp; Premereur et al.\u003csup\u003e13\u003c/sup\u003e), mapped area (28° × 28° vs 50° × 30°), or inherent heterogeneity within LO and across subjects. Both arrays showed contralateral receptive field biases, consistent with fMRI findings.\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e,\u003cspan additionalcitationids=\"CR25 CR26 CR27\" citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e–\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003cp\u003eAlthough limited to a single patient, our data provides robust evidence for size and positional invariance, the latter consistent across 2 distinct subregions of LO with different response profiles. This shows that these invariances are widespread within LO. Array 2 was positioned in a body patch,\u003csup\u003e29\u003c/sup\u003e as identified in previous work (P1 in Ramirez \u0026amp; Vanhoyland et al.\u003csup\u003e12\u003c/sup\u003e), and was tested for position and size invariance using headless torso stimuli. These contained detailed features, such as hands, unlike the simpler silhouettes used for array 1. Although these details could theoretically drive responses independently of shape contours, this seems unlikely, as a linear decoder trained on these torsos generalized to their silhouette counterparts (as shown in Ramirez \u0026amp; Vanhoyland et al.\u003csup\u003e12\u003c/sup\u003e), which is also consistent with macaque body-patch recordings.\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003cp\u003eNon-human primates have been extensively used to evaluate response invariance in both ventral and dorsal visual stream at the single-unit level.\u003csup\u003e\u003cspan additionalcitationids=\"CR2 CR3 CR4 CR5 CR6 CR7\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e–\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e,\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e,\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e\u003c/sup\u003e The inferotemporal cortex (ITC) in the ventral stream maintains shape selectivity despite positional shifts and exhibits robust size invariance.\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e Despite maintenance of the ranking order across transformations, the absolute magnitude of responses might change, similar to our observations.\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e,\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e However, the response pattern of multiple single-neurons is necessary in ITC to decode objects across changes in size and position.\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e Recent advancements in intracortical recordings in the human lateral occipital (LO) cortex reveal striking parallels with macaque ITC, such as similarities in coding of shape, stereopsis, bodies and faces.\u003csup\u003e\u003cspan additionalcitationids=\"CR10 CR11 CR12 CR13 CR14 CR15 CR16\" citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e–\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e Our findings corroborate these similarities, showing that human LO maintains stimulus selectivity across size changes of up to two octaves (the maximum that was tested), compared to tolerance of up to 4 octaves in macaque ITC.\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e Time-resolved correlational analysis revealed significant Pearson correlation coefficients of 0.29 for 2° vs. 4° (1 octave), 0.28 for 2° vs. 8° (2 octaves), and 0.40 for 4° vs. 8° (1 octave). The lower correlations for comparisons involving 2° stimuli likely stem from fewer neurons responding to these smaller shapes.\u003c/p\u003e\u003cp\u003eSingle-neuron recordings in human LO are lacking, and current evidence is based on fMRI and EEG. LO activity remains invariant to stimulus size changes of up to fourfold in fMRI,\u003csup\u003e18,21\u003c/sup\u003e and most studies support invariant stimulus representation across the visual field.\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e,\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e,\u003cspan additionalcitationids=\"CR34\" citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e–\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e BOLD signal adaptation occurs when stimuli are repeatedly presented with varying size or position, suggesting invariant neural representations, as intolerance would lead to activation of distinct neural populations, preventing BOLD suppression.\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e Additionally, both individual exemplars and superordinate categories are decodable across the visual field, even between hemifields.\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e,\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e,\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e It was hypothesizes that such location-tolerant coding could be explained by large receptive fields crossing the meridians,\u003csup\u003e28\u003c/sup\u003e based on the receptive fields from macaque ITC that typically range between 5 and 30°.\u003csup\u003e2,7,38–40\u003c/sup\u003e Our data, in line with Decramer \u0026amp; Premereur et al.\u003csup\u003e13\u003c/sup\u003e, supports this theory since a considerable number of neurons crossed the vertical or horizontal meridian. In contrast, one fMRI study found no evidence for individuation (ability to differentiate between individual objects based on the spatial response patterns) for objects across different positions within and between same hemifields.\u003csup\u003e\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e Notably, position-invariant category information extends beyond LO to most category-selective regions such as the extrastriate body area and fusiform face area.\u003csup\u003e\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u003c/sup\u003e Yet, LO appears uniquely capable of distinguishing stimulus location on cluttered backgrounds, with EEG recordings indicating a latency of 140–200 ms.\u003csup\u003e37\u003c/sup\u003e While fMRI adaptation paradigms aim to be less susceptible to the obscuring effects of averaging neural activity of potentially selective neural populations, all fMRI and EEG signals remain an ensemble of thousands of neurons simultaneously. Our single-neuron data therefore provide crucial validation of these neuroimaging findings, confirming that stimulus selectivity in human LO persists across size and position transformations. Moreover, the high temporal resolution of our recordings reveals that these invariances emerge during feedforward propagation, independent of top-down influences. These findings bridge a critical gap between non-human primate and human studies, offering direct evidence of single-neuron invariance in human LO.\u003c/p\u003e"},{"header":"Methods","content":"\u003ch2\u003eStudy approval\u003c/h2\u003e\u003cp\u003e We obtained ethical approval (study number s53126) for conducting semi-chronic microelectrode recordings using the Utah array in patients undergoing invasive epilepsy monitoring. The study protocol was approved by the local ethical committee (Ethische Commissie Onderzoek UZ/KU Leuven). The study was carried out in accordance with the principles of the Declaration of Helsinki, the principles of good clinical practice, and all applicable regulatory requirements set by the Federal Agency for Medicines and Health Products (FAGG).\u003c/p\u003e\u003cp\u003eTo ensure safety and compliance, strict adherence to imposed safety measures was maintained. This included the use of case report forms and detailed reports on any (serious) adverse events. All data collected during the study were encrypted and stored at the University Hospitals Leuven. The study was thoroughly discussed with the patient during the pre-operative consultation, which took place at least six weeks before the surgery. The patient was informed about the risks associated with micro-electrode array implantation, such as infection and hemorrhage. No additional incisions were made for the purpose of this study. Written informed consent was obtained from the patient on the evening prior to the surgery.\u003c/p\u003e\u003ch3\u003eClinical information\u003c/h3\u003e\u003cp\u003eWe obtained invasive intracranial recordings from 2 arrays from the left lateral occipital cortex in a patient with refractory epilepsy. She was a 58-year old female suffering from recurrent psychotic episodes. She was taking chronic anti-seizure medications (ASM) during admission: Levetiracetam 2000 mg, Phenytoin 100 mg, Clonazepam 1 mg and Lacosamide 300 mg, all twice daily. During admission to the epilepsy ward, the ASM was adjusted based on the number of detected seizures.\u003c/p\u003e\u003cp\u003eTarget locations of intracranial electrodes were determined by the epileptologist in a multidisciplinary consensus meeting and were based on electroclinical findings and non-invasive multimodal imaging. Multi-electrode arrays were implanted via the craniotomy performed for the implantation of a subdural grid, adjacent to the grid (Figure S4). The arrays were removed during the same surgery for subdural grid removal, 14 days after implantation. Both insertion and removal of electrodes were done under general anesthesia. No new neurological deficits, nor implantation-related complications (infection, hemorrhage, wound problems) were recorded (Figure S5).\u003c/p\u003e\u003cp\u003eAlthough the presumed epileptogenic zone encompassed the lateral occipital cortex based on multimodal preoperative assessment, no focal onset could be identified. Therefore, she was deemed not to be a good candidate for epilepsy surgery.\u003c/p\u003e\u003ch2\u003eInvasive recordings\u003c/h2\u003e\u003cp\u003eWe recorded from two 96-channel arrays (Utah Array™ – Blackrock Neurotech) from day 1 until the day before electrode removal. Arrays were inserted using a pneumatic inserter wand, provided by Blackrock Neurotech, with a single hit (20 psi). We closed the dura above the array and placed the bone flap on top to secure the array and prevent array flotation. Reference wires were placed subdurally and ground wires epidurally. The signal was digitally amplified using a Cereplex M head stage from Blackrock Neurotech, connected to a 128-channel neural signal processor (NeuroPort), and sampled using Central Software at a rate of 30 kHz. We applied a 750 Hz high-pass filter for spiking activity. The multi-unit detection threshold was placed at -3 standard deviations of the signal. The variable signal quality in the immediate postoperative period resulted in a difference in the number of responsive and selective channels during each recording session. Single units were manually extracted offline using Plexon Offline Sorter v4.7.1.\u003c/p\u003e\u003ch2\u003eImaging\u003c/h2\u003e\u003cp\u003eT1-weighted images were acquired using a 3-T MR scanner (Achieva dstream, Philips Medical Systems, Best, the Netherlands) at the University Hospitals Leuven for presurgical planning. The imaging parameters were as follows: time of echo 3.1 ms, inversion time 900 ms, slice thickness 1.0 mm, 400 × 400 matrix size, and 283 coronal slices. Following electrode placement, a CT scan (Siemens) was performed with a slice thickness of 1 mm, a voltage of 120 kV, and a dose length product of 819 mGy.cm. The CT scan was used to identify the exact electrode locations and rule out any hemorrhage. This postoperative CT was later fused with a postoperative MRI performed months after surgery to rule out structural alterations due to electrode insertion. No structural alterations (gliosis, ischemia, hemosiderin) were seen at the implantation sites (Figure S4).\u003csup\u003e\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003cp\u003eImaging preprocessing was done with SPM12 software (Wellcome Department of Cognitive Neurology, London, UK) running on MATLAB (Mathworks, Natick, MA, USA). The preprocessing steps included (1) setting the anterior commissure as the origin for both the MRI and CT scan (2) realigning the images, (3) co-registering the MRI and CT images and (4) warping the MRI and coregistered CT to a common high-resolution brain atlas (MNI152). Freesurfer was used to generate cortical 3D renderings of the subjects' pial surfaces.\u003csup\u003e\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e\u003c/sup\u003e The precise electrode location on the cortical surface and the MNI coordinates were determined using iElectrodes Software.\u003csup\u003e\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003ch2\u003eStimulus presentation\u003c/h2\u003e\u003cp\u003eWe conducted experiments in a dimmed hospital room using custom-built software for stimulus presentation. Stimuli were presented on a 60 Hz DELL-P2418HZM LED monitor, while the patient sat 60 cm away from the screen (1 pixel = 0.026°). Fixation was monitored using continuous eye-movement tracking (left or right eye, 1000 Hz, Eyelink 1000 plus in head free-to move mode). The patient was instructed to focus on a small red square (0.2 × 0.2°) positioned at the center of the display. If the gaze deviated from the electronically preset 3° by 3° fixation window, the trial was aborted. To synchronize the data, a photodiode was attached to the upper left corner of the screen, which detected a bright flash coinciding with the first frame of the stimulus. This flash was invisible to the patient. To accurately record baseline spike rate activity, an intertrial interval of at least 100 ms was introduced (i.e. the offset of the stimulus and the onset of fixation for the next trial).\u003c/p\u003e\u003ch2\u003eStimulus sets and experimental designs\u003c/h2\u003e\u003ch2\u003eReceptive field\u003c/h2\u003e\u003cp\u003eWe initially presented 10 images of headless torso’s \u003csup\u003e43,44\u003c/sup\u003e (size 4° by 4°) at the fixation marker while the patient passively fixated, and selected the torso that elicited the strongest response across all neurons from array 2, designating this as the \"preferred body\" (Figure S2). Stimulus presentation lasted 250 ms, and each trial was preceded by 300 ms of fixation. This was done on day 3 after implantation.\u003c/p\u003e\u003cp\u003eThis preferred body was then displayed at 49 (7 by 7) non-overlapping positions, thereby mapping the central 28° x 28°. Locations were mapped in randomized order. The patient performed a distractor task, pressing a button with her right hand whenever a cross appeared on the fixation marker, which occurred in approximately 2% of trials. Stimulus presentation lasted 250 ms, and each trial was preceded by 300 ms of fixation. Due to time constraints and the multi-electrode recording technique, we could not test the receptive field of each neuron individually with its optimal stimulus. The experiment was performed on day 12 after implantation.\u003c/p\u003e\u003ch2\u003eInvariance for position within receptive field\u003c/h2\u003e\u003cp\u003eWe used two stimulus sets to test for invariance due to the different selectivity profiles observed in both arrays. The stimuli were presented in different non-overlapping locations within the average receptive field for a particular array. Individual stimuli and on-screen location were presented in a randomized order.\u003c/p\u003e\u003cp\u003eThe first set consisted of 64 white shapes presented on a black background, of which 30% was filled with random black noise. Shapes were approximately 4° by 4°. All 64 shapes were presented 4° above or below the fixation marker (2 locations) with its center on the vertical meridian, thereby not crossing the horizontal meridian. Each shape was presented \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\pm\\:\\)\u003c/span\u003e\u003c/span\u003e 10 times in each location. Stimulus presentation lasted 250 ms, and each trial was preceded by 300 ms of fixation. This set was recorded in array 1 on day 11.\u003c/p\u003e\u003cp\u003eThe second set consisted of the same 10 torso’s as described for the receptive field testing (4° by 4°). All ten torsos were presented \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\pm\\:\\)\u003c/span\u003e\u003c/span\u003e 20 times in 5 locations in the lower right quadrant on a white background. The x- and y- coordinates of the center locations were: (4°,-8°),(8°,-8°), (12°,-8°), (8°,-12°), (12°,-12°). Stimulus presentation lasted 250 ms, and each trial was preceded by 300 ms of fixation. This set was recorded in array 2 on day 6.\u003c/p\u003e\u003cp\u003eIn both experiments, the patient performed a distractor task by pressing a button when a cross was presented on the fixation marker (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\pm\\:\\)\u003c/span\u003e\u003c/span\u003e 2% of trials).\u003c/p\u003e\u003ch2\u003eInvariance for changes in size\u003c/h2\u003e\u003cp\u003eThis experiment was only done for array 2. We presented 20 torso’s (10 of which were used for the position invariance experiment) in 3 different sizes (2° by 2°, 4° by 4° and 8° by 8°). All torsos were presented \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\pm\\:\\)\u003c/span\u003e\u003c/span\u003e 8 times in all 3 sizes on a white background. The torso’s were presented within the receptive field in the right lower quadrant with center location x = 10° and y = -10°. Stimulus presentation lasted 200 ms, and each trial was preceded by 300 ms of fixation. The patient was passively fixating. This set was recorded on day 4.\u003c/p\u003e\u003ch2\u003eAnalysis\u003c/h2\u003e\u003ch2\u003eData analysis\u003c/h2\u003e\u003cp\u003e We analyzed data using custom-written MATLAB R2023b (Mathworks, Natick, MA, USA) scripts. Net spike firing rates were calculated from raw multi-unit and single-unit firing rates by subtracting baseline activity (250 or 300 ms interval before stimulus onset) from the entire trial, binned in 10 ms intervals. Analyses were restricted to visually responsive sites. To assess the responsiveness of each recorded site, we performed two-sided dependent samples t-tests comparing spike rates during baseline and response windows for each stimulus category, condition or position separately. This approach ensured that sites with a strong preference for a specific category or stimulus are not overlooked by a broad test across all stimuli. For each experiment, a site was considered visually responsive if any condition showed a significant difference from baseline (p \u0026lt; 0.005 to account for multiple testing) in the 150 ms window following the response latency, the ‘response window’ (Array 1: 130–280 ms; Array 2: 200–350 ms). Response windows have been reported in a previous research article (array 1 corresponds to array 3, and array 2 to array 4).\u003csup\u003e45\u003c/sup\u003e\u003c/p\u003e\u003ch2\u003eData normalization\u003c/h2\u003e\u003cp\u003eAll subsequent analyses utilized baseline-subtracted average net firing rates or z-normalized net firing rates. Z-normalization was performed for each 10 ms-binned multi-unit or single-unit recording as follows:\u003c/p\u003e\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:Z\\left(i\\right)=\\:\\frac{R\\left(i\\right)-{R}_{mean}}{{R}_{std}}$$\u003c/div\u003e\u003c/div\u003e\u003cp\u003ewhere \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:Z\\left(i\\right)\\)\u003c/span\u003e\u003c/span\u003e corresponds to the Z-normalized net spike rate for a specific 10ms bin, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:R\\left(i\\right)\\)\u003c/span\u003e\u003c/span\u003e to the average net response from the same 10 ms bin, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}_{mean}\\)\u003c/span\u003e\u003c/span\u003e is the average net spike rate over all trials in the response window, and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}_{std}\\)\u003c/span\u003e\u003c/span\u003e is the standard deviation in the response window. Within the experiments testing for invariance, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}_{mean}\\)\u003c/span\u003e\u003c/span\u003e and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{R}_{std}\\)\u003c/span\u003e\u003c/span\u003e were taken from all stimuli shown in the same position (for position invariance) or same size (for size invariance). This method ensures that the firing rates are standardized, allowing for consistent comparisons across different conditions and sites.\u003c/p\u003e\u003ch2\u003eReceptive field\u003c/h2\u003e\u003cp\u003eWe calculated the average response across all 49 locations within the response window, selecting only units that were responsive in at least one location with a minimum net spike rate of 5 spikes/s (p \u0026lt; 0.005, unpaired t-test comparing the response window to baseline, corrected for multiple testing). The location with the highest net response was identified as the center of the receptive field. Other locations were included in the receptive field if their response exceeded 50% of the center response and was significantly higher than baseline (p \u0026lt; 0.05, one-sided dependent samples t-test).\u003c/p\u003e\u003cp\u003eThe receptive field size was determined as the number of locations with a response ≥ 50% of the center response. To assess laterality (contralateral, ipsilateral, or bilateral) and verticality (upper hemifield, lower hemifield, or both), locations on the vertical and horizontal meridians were excluded, as these spanned the meridians.\u003c/p\u003e\u003cp\u003eTo generate an average receptive field per array, the net responses at each location were scaled per unit by dividing by the center response (thereby, center response equals 1). An average was then computed across all responsive units. For visualization, a Gaussian filter (σ = 0.5) was applied, followed by 2D cubic interpolation (refinement factor = 8).\u003c/p\u003e\u003ch2\u003eShape-tuning index and depth of selectivity\u003c/h2\u003e\u003cp\u003eBoth metrics were calculated on the position invariance experiment within the ‘preferred location’ (further explained in the Pearson correlation analysis) for all visually responsive neurons. The shape-tuning index quantifies the degree of shape tuning and is defined as \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\frac{MAX\\:-\\:MIN}{MAX}\\)\u003c/span\u003e\u003c/span\u003e, in which ‘\u003cem\u003eMAX’\u003c/em\u003e and \u003cem\u003e‘MIN’\u003c/em\u003e are the net spike rate within the response window to the most and least preferred stimulus at the ‘preferred’ location, respectively. A value of 0 indicates no shape-tuning, whereas a value of 1 (or even \u0026gt; 1) indicates strong shape-tuning. The depth of selectivity was calculated as \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\frac{N-\\:\\frac{SUM}{MAX}}{N-1}\\)\u003c/span\u003e\u003c/span\u003e, where N = number of stimuli presented at the preferred location, SUM = sum of the average net spike rate within the response window to all stimuli, and MAX = net spike within the response window to the preferred stimulus at the preferred location. The depth of selectivity is 1 if the neuron responds to only 1 stimulus, and 0 if it responds to all stimuli equally.\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e\u003ch2\u003ePearson correlation analysis\u003c/h2\u003e\u003cp\u003eTo quantify neural information in a time-resolved manner, firing rates were computed within 100 ms bins using a 10 ms sliding window. For each visually responsive single unit, the net response to each stimulus at each position (or size) was calculated within each time bin. Single-units that were only responsive in \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\ge\\:\\:\\)\u003c/span\u003e\u003c/span\u003e1 location for any stimulus (p \u0026lt; 0.005, dependent samples t-test comparing response window vs. baseline) were excluded from the analysis. For each single unit, the position (or size) eliciting the strongest response to any stimulus within each time bin was identified as the \"preferred location/size\" for that particular time bin. Stimuli were then ranked according to their z-normalized responses at the preferred location/size. For the 64 shapes presented in array 1, besides using all stimuli, we also used the 5 best and 5 worst stimuli only to make it more comparable to the 10 torso’s from array 2.\u003c/p\u003e\u003cp\u003eWe first estimated the noise ceiling, the maximal correlation that is theoretically obtainable. For this, we randomly split for each neuron the trials belonging to the preferred stimuli (which could be different for each single unit) at the preferred location/size in 2 equally sized groups. The average of both groups was taken, creating response \"pairs\" as a measure of intertrial variability (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{N}_{pairs\\:}={N}_{single\\:units\\:}*{N}_{stimuli})\\)\u003c/span\u003e\u003c/span\u003e. This process was iterated 1000 times for each time bin, and for each iteration we calculated the Pearson correlation coefficient (r) over all response pairs using the \u003cem\u003ecorrcoef\u003c/em\u003e function. Pearson’s r was averaged, and a 95% confidence interval was obtained from the null distribution of Pearson coefficients.\u003c/p\u003e\u003cp\u003eTo estimate the response invariance within the receptive field, the z-normalized responses from all other visually responsive positions/sizes (in which \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\ge\\:\\:\\)\u003c/span\u003e\u003c/span\u003e 1 stimulus was visually responsive, p \u0026lt; 0.05, two-sided dependent samples t-test comparing response window vs. baseline), termed the \"non-preferred positions/sizes\", were sorted according to the ranking obtained from the preferred position/size, creating response \"pairs\" between the preferred and non-preferred positions/sizes. These response pairs, pooled across all visually responsive neurons, were used to compute Pearson’s r for each time bin. We performed Fisher’s z-transformation to derive 95% confidence intervals [\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{r}_{lower\\:boundary}\\:{r}_{upper\\:boundary}]\\)\u003c/span\u003e\u003c/span\u003e of the Pearson coefficient r as follows:\u003c/p\u003e\u003cdiv id=\"Equb\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equb\" name=\"EquationSource\"\u003e\n$$\\:{Z}_{lower\\:boundary}^{upper\\:boundary}=0.5*\\text{ln}\\left(\\frac{1+r}{1-r}\\right)\\pm\\:\\left(1.96*\\frac{1}{\\sqrt{N-3}}\\right)\\:\\:with\\:N={N}_{single\\:units}*{N}_{stimuli}$$\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equc\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equc\" name=\"EquationSource\"\u003e\n$$\\:{r}_{lower\\:boundary}^{upper\\:boundary}=\\:\\frac{{e}^{2Z\\:}-1}{{e}^{2Z}+1}\\:wih\\:Z={Z}_{upper\\:boundary}\\:or\\:{Z}_{lower\\:boundary}$$\u003c/div\u003e\u003c/div\u003e\u003cp\u003eSignificance was assessed with the associated two-tailed t-test of the null hypothesis that the population correlation coefficient equals zero (df = n – 2).\u003c/p\u003e\u003ch2\u003eOrdinary Least Squares (OLS) regression analysis\u003c/h2\u003e\u003cp\u003eWe fitted a linear regression through the sorted (from high to low) responses, per single unit,\u003c/p\u003e\u003cp\u003efrom the non-preferred positions/sizes using the Ordinary Least Squares method. We calculated the average slope and 95% confidence intervals. A negative regression slope (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\beta\\:\\)\u003c/span\u003e\u003c/span\u003e) significantly different from zero signifies response invariance to some degree between different locations. The reported slope equals the calculated slope, multiplied by the number of stimuli (64 for array 1; 10 for array 2) to be able to compare slopes across arrays.\u003c/p\u003e\u003ch2\u003eMulti-dimensional scaling\u003c/h2\u003e\u003cp\u003eTo visualize the similarity relationships in the neuronal population between the torso’s/shapes and the position within the receptive field, we employed multidimensional scaling (MDS). The neuronal distances were computed as the average Euclidean distance of the baseline-subtracted average Z-normalized firing rates across neurons for each image pair using MATLAB’s \u003cem\u003epdist\u003c/em\u003e function. This way we generated a dissimilarity matrix, with both dimensions corresponding to the number of individual images, that held the Euclidian distances measured in multidimensional space (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{N}_{dimensions\\:}=\\:\\)\u003c/span\u003e\u003c/span\u003enumber of visually responsive single-units). The MDS procedure using MATLAB’s \u003cem\u003emdscale\u003c/em\u003e function was applied to the dissimilarity matrix of each experiment. To evaluate the alignment between the two-dimensional embedding and the actual multi-dimensional neuronal distances, we calculated the Pearson correlation between them. The resulting correlation coefficient is displayed in each multidimensional scaling plot. High correlation values indicate that the multidimensional scaling plots provide a reasonable approximation of the underlying similarity relationships. Given that we observed high correlations, we calculated the average distance (in this two-dimensional embedding) \u003cem\u003e‘within’\u003c/em\u003e and \u003cem\u003e‘between’\u003c/em\u003e groups in 100ms bins with a 10ms sliding window (e.g. ‘\u003cem\u003ewithin’\u003c/em\u003e a certain location in the receptive field and ‘\u003cem\u003ebetween\u003c/em\u003e’ different locations). To calculate the average Euclidian distance, we determined, for each stimulus individually, the distance to all stimuli within the same group (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{N}_{samples\\:within\\:group}=\\:n*\\:\\left((\\sum\\:_{k=1}^{s}k)\\right.-\\:\\left.s\\right)\\:)\\)\u003c/span\u003e\u003c/span\u003e and to all stimuli from all others groups (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:{N}_{samples\\:between\\:group}=s*\\left(s-1\\right)*(\\left(\\sum\\:_{k=1}^{n}k\\right)-n)\\)\u003c/span\u003e\u003c/span\u003e ) with \u003cem\u003en = number of groups\u003c/em\u003e and \u003cem\u003es = number of stimuli per group\u003c/em\u003e.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003ch2\u003eDeclaration of Interests\u003c/h2\u003e\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e\u003cp\u003eThis work was supported by Fonds Wetenschappelijk Onderzoek (FWO) grant G0B6422N and KU Leuven grant C14/22/134. T.T. is supported by FWO (senior clinical researcher; FWO 1830717N). M.V. holds an FWO fellowship for fundamental research (1169321N).\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eM.V. conceived and designed the experiments. T.T. and M.V. planned and performed array implantations. M.V. performed the recordings. M.V. performed all clinical trial related activities. M.V. analyzed the data and wrote the manuscript. T.T. and P.J. supervised and guided the study. All authors reviewed and edited the manuscript.\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eWe are grateful to the patient for her participation. We thank Stijn Verstraeten and Wouter Depuydt for technical assistance with the recording setup, Ana\u0026iuml;s Van Hoylandt for assisting with data recording and data monitoring and Thomas Decramer for fine-tuning the implantation and recording technique from Utah-arrays in humans. We thank Rufin Vogels for providing the simple shape stimuli used in array 1.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe raw data is uploaded to a figshare repository at https://doi.org/10.6084/m9.figshare.30337921.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eIto, M., Tamura, H., Fujita, I. \u0026amp; Tanaka, K. Size and position invariance of neuronal responses in monkey inferotemporal cortex. \u003cem\u003eJ. Neurophysiol.\u003c/em\u003e \u003cb\u003e73\u003c/b\u003e, 218\u0026ndash;226. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1152/jn.1995.73.1.218\u003c/span\u003e\u003cspan address=\"10.1152/jn.1995.73.1.218\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (1995).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDe Op, H. \u0026amp; Vogels, R. Spatial sensitivity of macaque inferior temporal neurons. \u003cem\u003eJ. Comp. Neurol.\u003c/em\u003e \u003cb\u003e426\u003c/b\u003e, 505\u0026ndash;518. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/1096-9861(20001030)426:4%3C505::aid-cne1%3E3.0.co;2-m\u003c/span\u003e\u003cspan address=\"10.1002/1096-9861(20001030)426:4%3C505::aid-cne1%3E3.0.co;2-m\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2000).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSereno, A. B. \u0026amp; Maunsell, J. H. Shape selectivity in primate lateral intraparietal cortex. \u003cem\u003eNature\u003c/em\u003e \u003cb\u003e395\u003c/b\u003e, 500\u0026ndash;503. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/26752\u003c/span\u003e\u003cspan address=\"10.1038/26752\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (1998).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSary, G., Vogels, R. \u0026amp; Orban, G. A. Cue-invariant shape selectivity of macaque inferior temporal neurons. \u003cem\u003eScience\u003c/em\u003e \u003cb\u003e260\u003c/b\u003e, 995\u0026ndash;997. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1126/science.8493538\u003c/span\u003e\u003cspan address=\"10.1126/science.8493538\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (1993).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLogothetis, N. K., Pauls, J. \u0026amp; Poggio, T. Shape representation in the inferior temporal cortex of monkeys. \u003cem\u003eCurr. Biol.\u003c/em\u003e \u003cb\u003e5\u003c/b\u003e, 552\u0026ndash;563. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/s0960-9822(95)00108-4\u003c/span\u003e\u003cspan address=\"10.1016/s0960-9822(95)00108-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (1995).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJanssen, P., Srivastava, S., Ombelet, S. \u0026amp; Orban, G. A. Coding of shape and position in macaque lateral intraparietal area. \u003cem\u003eJ. Neurosci.\u003c/em\u003e \u003cb\u003e28\u003c/b\u003e, 6679\u0026ndash;6690. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1523/JNEUROSCI.0499-08.2008\u003c/span\u003e\u003cspan address=\"10.1523/JNEUROSCI.0499-08.2008\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2008).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDiCarlo, J. J. \u0026amp; Maunsell, J. H. Anterior inferotemporal neurons of monkeys engaged in object recognition can be highly sensitive to object retinal position. \u003cem\u003eJ. Neurophysiol.\u003c/em\u003e \u003cb\u003e89\u003c/b\u003e, 3264\u0026ndash;3278. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1152/jn.00358.2002\u003c/span\u003e\u003cspan address=\"10.1152/jn.00358.2002\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2003).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYamane, Y., Carlson, E. T., Bowman, K. C., Wang, Z. \u0026amp; Connor, C. E. A neural code for three-dimensional object shape in macaque inferotemporal cortex. \u003cem\u003eNat. Neurosci.\u003c/em\u003e \u003cb\u003e11\u003c/b\u003e, 1352\u0026ndash;1360. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/nn.2202\u003c/span\u003e\u003cspan address=\"10.1038/nn.2202\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2008).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVerhoef, B. E., Vogels, R. \u0026amp; Janssen, P. Contribution of inferior temporal and posterior parietal activity to three-dimensional shape perception. \u003cem\u003eCurr. Biol.\u003c/em\u003e \u003cb\u003e20\u003c/b\u003e, 909\u0026ndash;913. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.cub.2010.03.058\u003c/span\u003e\u003cspan address=\"10.1016/j.cub.2010.03.058\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2010).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVerhoef, B. E., Vogels, R. \u0026amp; Janssen, P. Inferotemporal cortex subserves three-dimensional structure categorization. \u003cem\u003eNeuron\u003c/em\u003e \u003cb\u003e73\u003c/b\u003e, 171\u0026ndash;182. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.neuron.2011.10.031\u003c/span\u003e\u003cspan address=\"10.1016/j.neuron.2011.10.031\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2012).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJanssen, P., Vogels, R., Liu, Y. \u0026amp; Orban, G. A. At least at the level of inferior temporal cortex, the stereo correspondence problem is solved. \u003cem\u003eNeuron\u003c/em\u003e \u003cb\u003e37\u003c/b\u003e, 693\u0026ndash;701. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/s0896-6273(03)00023-0\u003c/span\u003e\u003cspan address=\"10.1016/s0896-6273(03)00023-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2003).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRamirez, J. G. et al. Intracortical recordings reveal the neuronal selectivity for bodies and body parts in the human visual cortex. \u003cem\u003eProc. Natl. Acad. Sci. U S A\u003c/em\u003e. \u003cb\u003e121\u003c/b\u003e, e2408871121. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1073/pnas.2408871121\u003c/span\u003e\u003cspan address=\"10.1073/pnas.2408871121\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDecramer, T. et al. Single-cell selectivity and functional architecture of human lateral occipital complex. \u003cem\u003ePLoS Biol.\u003c/em\u003e \u003cb\u003e17\u003c/b\u003e, e3000280. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pbio.3000280\u003c/span\u003e\u003cspan address=\"10.1371/journal.pbio.3000280\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDecramer, T. et al. Single-Unit Recordings Reveal the Selectivity of a Human Face Area. \u003cem\u003eJ. Neurosci.\u003c/em\u003e \u003cb\u003e41\u003c/b\u003e, 9340\u0026ndash;9349. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1523/JNEUROSCI.0349-21.2021\u003c/span\u003e\u003cspan address=\"10.1523/JNEUROSCI.0349-21.2021\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBougou, V. et al. Neuronal tuning and population representations of shape and category in human visual cortex. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cb\u003e15\u003c/b\u003e, 4608. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41467-024-49078-3\u003c/span\u003e\u003cspan address=\"10.1038/s41467-024-49078-3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVogels, R. More Than the Face: Representations of Bodies in the Inferior Temporal Cortex. \u003cem\u003eAnnu. Rev. Vis. Sci.\u003c/em\u003e \u003cb\u003e8\u003c/b\u003e, 383\u0026ndash;405. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1146/annurev-vision-100720-113429\u003c/span\u003e\u003cspan address=\"10.1146/annurev-vision-100720-113429\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAfraz, S. R., Kiani, R. \u0026amp; Esteky, H. Microstimulation of inferotemporal cortex influences face categorization. \u003cem\u003eNature\u003c/em\u003e \u003cb\u003e442\u003c/b\u003e, 692\u0026ndash;695. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/nature04982\u003c/span\u003e\u003cspan address=\"10.1038/nature04982\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2006).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMalach, R. et al. Object-related activity revealed by functional magnetic resonance imaging in human occipital cortex. \u003cem\u003eProc. Natl. Acad. Sci. U S A\u003c/em\u003e. \u003cb\u003e92\u003c/b\u003e, 8135\u0026ndash;8139. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1073/pnas.92.18.8135\u003c/span\u003e\u003cspan address=\"10.1073/pnas.92.18.8135\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (1995).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKriegeskorte, N. et al. Matching categorical object representations in inferior temporal cortex of man and monkey. \u003cem\u003eNeuron\u003c/em\u003e \u003cb\u003e60\u003c/b\u003e, 1126\u0026ndash;1141. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.neuron.2008.10.043\u003c/span\u003e\u003cspan address=\"10.1016/j.neuron.2008.10.043\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2008).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGrill-Spector, K. K. \u0026amp; Kanwisher, Z. The lateral occipital complex and its role in object recognition. \u003cem\u003eVis. Res.\u003c/em\u003e \u003cb\u003e41\u003c/b\u003e, 1409\u0026ndash;1422 (2001).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGrill-Spector, K. et al. Differential processing of objects under various viewing conditions in the human lateral occipital complex. \u003cem\u003eNeuron\u003c/em\u003e \u003cb\u003e24\u003c/b\u003e, 187\u0026ndash;203. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/s0896-6273(00)80832-6\u003c/span\u003e\u003cspan address=\"10.1016/s0896-6273(00)80832-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (1999).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSayres, R. \u0026amp; Grill-Spector, K. Relating retinotopic and object-selective responses in human lateral occipital cortex. \u003cem\u003eJ. Neurophysiol.\u003c/em\u003e \u003cb\u003e100\u003c/b\u003e, 249\u0026ndash;267. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1152/jn.01383.2007\u003c/span\u003e\u003cspan address=\"10.1152/jn.01383.2007\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2008).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRainer, G., Asaad, W. F. \u0026amp; Miller, E. K. Selective representation of relevant information by neurons in the primate prefrontal cortex. \u003cem\u003eNature\u003c/em\u003e \u003cb\u003e393\u003c/b\u003e, 577\u0026ndash;579. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/31235\u003c/span\u003e\u003cspan address=\"10.1038/31235\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (1998).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCichy, R. M., Chen, Y. \u0026amp; Haynes, J. D. Encoding the identity and location of objects in human LOC. \u003cem\u003eNeuroimage\u003c/em\u003e \u003cb\u003e54\u003c/b\u003e, 2297\u0026ndash;2307. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.neuroimage.2010.09.044\u003c/span\u003e\u003cspan address=\"10.1016/j.neuroimage.2010.09.044\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2011).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKravitz, D. J., Kriegeskorte, N. \u0026amp; Baker, C. I. High-level visual object representations are constrained by position. \u003cem\u003eCereb. Cortex\u003c/em\u003e. \u003cb\u003e20\u003c/b\u003e, 2916\u0026ndash;2925. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/cercor/bhq042\u003c/span\u003e\u003cspan address=\"10.1093/cercor/bhq042\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2010).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHemond, C. C. \u0026amp; Kanwisher, N. G. Op de Beeck, H. P. A preference for contralateral stimuli in human object- and face-selective cortex. \u003cem\u003ePLoS One\u003c/em\u003e. \u003cb\u003e2\u003c/b\u003e, e574. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pone.0000574\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0000574\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2007).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMacEvoy, S. P. \u0026amp; Epstein, R. A. Position selectivity in scene- and object-responsive occipitotemporal regions. \u003cem\u003eJ. Neurophysiol.\u003c/em\u003e \u003cb\u003e98\u003c/b\u003e, 2089\u0026ndash;2098. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1152/jn.00438.2007\u003c/span\u003e\u003cspan address=\"10.1152/jn.00438.2007\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2007).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eCichy, R. M. et al. Probing principles of large-scale object representation: category preference and location encoding. \u003cem\u003eHum. Brain Mapp.\u003c/em\u003e \u003cb\u003e34\u003c/b\u003e, 1636\u0026ndash;1651. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/hbm.22020\u003c/span\u003e\u003cspan address=\"10.1002/hbm.22020\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2013).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDowning, P. E., Jiang, Y., Shuman, M. \u0026amp; Kanwisher, N. A cortical area selective for visual processing of the human body. \u003cem\u003eScience\u003c/em\u003e \u003cb\u003e293\u003c/b\u003e, 2470\u0026ndash;2473. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1126/science.1063414\u003c/span\u003e\u003cspan address=\"10.1126/science.1063414\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2001).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePopivanov, I. D., Jastorff, J., Vanduffel, W. \u0026amp; Vogels, R. Tolerance of macaque middle STS body patch neurons to shape-preserving stimulus transformations. \u003cem\u003eJ. Cogn. Neurosci.\u003c/em\u003e \u003cb\u003e27\u003c/b\u003e, 1001\u0026ndash;1016. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1162/jocn_a_00762\u003c/span\u003e\u003cspan address=\"10.1162/jocn_a_00762\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2015).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHung, C. P., Kreiman, G., Poggio, T. \u0026amp; DiCarlo, J. J. Fast readout of object identity from macaque inferior temporal cortex. \u003cem\u003eScience\u003c/em\u003e \u003cb\u003e310\u003c/b\u003e, 863\u0026ndash;866. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1126/science.1117593\u003c/span\u003e\u003cspan address=\"10.1126/science.1117593\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2005).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRust, N. C. \u0026amp; Dicarlo, J. J. Selectivity and tolerance (invariance) both increase as visual information propagates from cortical area V4 to IT. \u003cem\u003eJ. Neurosci.\u003c/em\u003e \u003cb\u003e30\u003c/b\u003e, 12978\u0026ndash;12995. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1523/JNEUROSCI.0179-10.2010\u003c/span\u003e\u003cspan address=\"10.1523/JNEUROSCI.0179-10.2010\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2010).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTootell, R. B., Mendola, J. D., Hadjikhani, N. K., Liu, A. K. \u0026amp; Dale, A. M. The representation of the ipsilateral visual field in human cerebral cortex. \u003cem\u003eProc. Natl. Acad. Sci. U S A\u003c/em\u003e. \u003cb\u003e95\u003c/b\u003e, 818\u0026ndash;824. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1073/pnas.95.3.818\u003c/span\u003e\u003cspan address=\"10.1073/pnas.95.3.818\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (1998).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGrill-Spector, K.et al. A sequence of object-processing stages revealed by fMRI in the human occipital lobe.\u003cem\u003eHum Brain Mapp\u003c/em\u003e 6, 316\u0026ndash;328 (1998). https://doi.org/10.1002/(SICI)1097-0193(1998)6:4\u0026lt;316::AID-HBM9\u0026gt;3.0.CO;2-6\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSchwarzlose, R. F., Swisher, J. D., Dang, S. \u0026amp; Kanwisher, N. The distribution of category and location information across object-selective regions in human visual cortex. \u003cem\u003eProc. Natl. Acad. Sci. U S A\u003c/em\u003e. \u003cb\u003e105\u003c/b\u003e, 4447\u0026ndash;4452. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1073/pnas.0800431105\u003c/span\u003e\u003cspan address=\"10.1073/pnas.0800431105\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2008).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWilliams, M. A. et al. Feedback of visual object information to foveal retinotopic cortex. \u003cem\u003eNat. Neurosci.\u003c/em\u003e \u003cb\u003e11\u003c/b\u003e, 1439\u0026ndash;1445. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/nn.2218\u003c/span\u003e\u003cspan address=\"10.1038/nn.2218\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2008).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGraumann, M., Ciuffi, C., Dwivedi, K., Roig, G. \u0026amp; Cichy, R. M. The spatiotemporal neural dynamics of object location representations in the human brain. \u003cem\u003eNat. Hum. Behav.\u003c/em\u003e \u003cb\u003e6\u003c/b\u003e, 796\u0026ndash;811. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41562-022-01302-0\u003c/span\u003e\u003cspan address=\"10.1038/s41562-022-01302-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDesimone, R., Albright, T. D., Gross, C. G. \u0026amp; Bruce, C. Stimulus-selective properties of inferior temporal neurons in the macaque. \u003cem\u003eJ. Neurosci.\u003c/em\u003e \u003cb\u003e4\u003c/b\u003e, 2051\u0026ndash;2062. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1523/JNEUROSCI.04-08-02051.1984\u003c/span\u003e\u003cspan address=\"10.1523/JNEUROSCI.04-08-02051.1984\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (1984).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKobatake, E. \u0026amp; Tanaka, K. Neuronal selectivities to complex object features in the ventral visual pathway of the macaque cerebral cortex. \u003cem\u003eJ. Neurophysiol.\u003c/em\u003e \u003cb\u003e71\u003c/b\u003e, 856\u0026ndash;867. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1152/jn.1994.71.3.856\u003c/span\u003e\u003cspan address=\"10.1152/jn.1994.71.3.856\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (1994).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eRichmond, B. J., Wurtz, R. H. \u0026amp; Sato, T. Visual responses of inferior temporal neurons in awake rhesus monkey. \u003cem\u003eJ. Neurophysiol.\u003c/em\u003e \u003cb\u003e50\u003c/b\u003e, 1415\u0026ndash;1432. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1152/jn.1983.50.6.1415\u003c/span\u003e\u003cspan address=\"10.1152/jn.1983.50.6.1415\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (1983).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDale, A. M., Fischl, B. \u0026amp; Sereno, M. I. Cortical surface-based analysis. I. Segmentation and surface reconstruction. \u003cem\u003eNeuroimage\u003c/em\u003e \u003cb\u003e9\u003c/b\u003e, 179\u0026ndash;194. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1006/nimg.1998.0395\u003c/span\u003e\u003cspan address=\"10.1006/nimg.1998.0395\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (1999).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBlenkmann, A. O. et al. iElectrodes: A Comprehensive Open-Source Toolbox for Depth and Subdural Grid Electrode Localization. \u003cem\u003eFront. Neuroinform\u003c/em\u003e. \u003cb\u003e11\u003c/b\u003e, 14. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/fninf.2017.00014\u003c/span\u003e\u003cspan address=\"10.3389/fninf.2017.00014\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2017).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePopivanov, I. D., Jastorff, J., Vanduffel, W. \u0026amp; Vogels, R. Stimulus representations in body-selective regions of the macaque cortex assessed with event-related fMRI. \u003cem\u003eNeuroimage\u003c/em\u003e \u003cb\u003e63\u003c/b\u003e, 723\u0026ndash;741. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.neuroimage.2012.07.013\u003c/span\u003e\u003cspan address=\"10.1016/j.neuroimage.2012.07.013\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2012).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePeelen, M. V. \u0026amp; Downing, P. E. Selectivity for the human body in the fusiform gyrus. \u003cem\u003eJ. Neurophysiol.\u003c/em\u003e \u003cb\u003e93\u003c/b\u003e, 603\u0026ndash;608. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1152/jn.00513.2004\u003c/span\u003e\u003cspan address=\"10.1152/jn.00513.2004\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2005).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVanhoyland, M., Janssen, P. \u0026amp; Theys, T. Single-neuron activity reflects visual awareness in human lateral occipital complex. \u003cem\u003ePreprint Res. Square\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.21203/rs.3.rs-5941203/v1\u003c/span\u003e\u003cspan address=\"10.21203/rs.3.rs-5941203/v1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2025). https://doi.org/.\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eVanhoyland, M., Janssen, P. \u0026amp; Theys, T. Data from Human lateral occipital complex is invariant for position and size transformations at the single-neuron level \u003cem\u003efigshare\u003c/em\u003e (2025). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.6084/m9.figshare.30337921\u003c/span\u003e\u003cspan address=\"10.6084/m9.figshare.30337921\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\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":"","lastPublishedDoi":"10.21203/rs.3.rs-8222241/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8222241/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eObject recognition relies on neural invariance to changes in position and size, yet the underlying mechanisms in humans remain uncertain due to the challenges that accompany single-unit recordings in cortical areas. We conducted the first single-unit recordings investigating position and size invariance in the human lateral occipital complex (LO) using two microelectrode arrays implanted in distinct LO subregions of a single patient. Our findings reveal that LO neurons exhibit strong shape tuning and have smaller receptive fields than previously reported. Time-resolved correlational analyses and multidimensional scaling confirmed robust position invariance and size invariance for changes up to 2 octaves, mirroring properties of the macaque inferotemporal cortex. These results advance our understanding of LO\u0026rsquo;s critical role in visual processing and object recognition.\u003c/p\u003e","manuscriptTitle":"Human lateral occipital complex is invariant for position and size transformations at the single-neuron level","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-12-08 07:31:44","doi":"10.21203/rs.3.rs-8222241/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-01-28T23:49:14+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-01-28T10:58:43+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"216952310407042966914021589254991917521","date":"2026-01-12T11:36:15+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-12-23T21:31:28+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"192006901984474685509613102408497999541","date":"2025-12-03T18:43:34+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-12-03T08:21:19+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-12-03T07:59:10+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-12-02T20:29:46+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-12-01T20:52:43+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-12-01T20:46:32+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":"42d1a826-3559-4ae8-a08d-6cd257b450ea","owner":[],"postedDate":"December 8th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":59112672,"name":"Biological sciences/Biological techniques"},{"id":59112673,"name":"Biological sciences/Neuroscience"}],"tags":[],"updatedAt":"2026-03-16T16:11:16+00:00","versionOfRecord":{"articleIdentity":"rs-8222241","link":"https://doi.org/10.1038/s41598-026-43946-2","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2026-03-12 16:00:21","publishedOnDateReadable":"March 12th, 2026"},"versionCreatedAt":"2025-12-08 07:31:44","video":"","vorDoi":"10.1038/s41598-026-43946-2","vorDoiUrl":"https://doi.org/10.1038/s41598-026-43946-2","workflowStages":[]},"version":"v1","identity":"rs-8222241","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8222241","identity":"rs-8222241","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}