Temporal Lobe Activation Predicts Episodic Memory Following Traumatic Brain Injury

Temporal Lobe Activation Predicts Episodic Memory Following Traumatic Brain Injury | 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 Research Article Temporal Lobe Activation Predicts Episodic Memory Following Traumatic Brain Injury Abbie Taing, Matthew Mundy, Jennie Ponsford, Gershon Spitz This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-152808/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 8 You are reading this latest preprint version Abstract The temporal lobes are critical for episodic memories and are preferentially affected following a traumatic brain injury (TBI). As such, episodic memory difficulties are common following TBI; however, the underlying neural changes that precipitate or maintain these difficulties in the early phase of recovery remains poorly understood. Here, we use functional magnetic resonance imaging (fMRI) to interrogate the relationship of temporal lobe activation in response to face, scene, and animal stimuli. Twenty-five patients with moderate to severe TBI were recruited an average of 2 months’ post-injury and compared with 21 demographically similar healthy controls. Findings indicate that memory for faces was preferentially impaired, compared to scene and animal stimuli. Decreased activity in temporal lobe structures was present for both face (right transverse temporal gyrus) and scene stimuli (right fusiform gyrus), but not for animals. Greater activation in these structures was associated with better long-term recognition. These findings provide evidence to suggest that TBI: a) preferentially affects memory for complex stimuli such as faces and scenes, and b) causes aberrant neuronal processes despite lack of evidence of significant impairment in behavioural performance. The mechanisms underpinning these findings are discussed in terms of differences in verbalisation during encoding and reduced neural efficiency. Psychology Cognitive Neuroscience traumatic brain injury episodic memory MRI temporal lobes Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Traumatic brain injury (TBI) can cause focal and diffuse disruption to multiple brain systems. Pathology is most widely observed in frontal and temporal cortices 1 , and as such impairments in processing speed, attention, executive function, and memory are most common 2,3 . Invariably, all individuals with a moderate or severe TBI will experience an initial transient period of impaired consciousness with amnesia and general confusion known as post-traumatic amnesia (PTA) 4 . Although most individuals emerge from PTA, many report ongoing difficulty with so-called episodic memories 5,6 . Episodic memories involve the ability to learn, store, and retrieve information about personal experiences 7,8 . Impairment of episodic memory can interfere with crucial skills such as new learning and task completion, and therefore can significantly limit functional independence and productivity 9 . Neuroanatomically, episodic memory is supported by a brain network spanning several regions in the frontal and parietal cortices 10,11 . However, structures contained within the temporal lobes are arguably the most crucial for particular aspects of episodic memory 7,12 . The temporal lobes house the hippocampus, and the entorhinal, perirhinal, and parahippocampal cortices – collectively known as the medial temporal lobes 13,14 . The hippocampus and other medial temporal structures are involved in the encoding of episodic stimuli, and modulation of activity in these regions has been found to predict subsequent recollection of the encoded stimuli 15 . The temporal lobes are preferentially affected after TBI 1 . The location of the temporal lobes and their proximity to the middle cranial fossa makes them highly susceptible to acceleration-deceleration forces present during injury 16,17 . Structures within the temporal lobes, such as the hippocampus and other limbic structures, are particularly susceptible to hypoxic insults and excitotoxicity effects 12 . As such, atrophy in the hippocampus and fornix is common following TBI 18 , and has been associated with impaired memory 19 . Functional imaging studies in the TBI population have also implicated the temporal lobes in impaired episodic memory. Individuals with TBI tend to display increased activity in various regions, including the temporal lobes, during encoding of episodic stimuli when compared with healthy controls 20-22 . One limitation of these previous studies is that most have failed to use paradigms that may reflect the type of episodic memory deficits that individuals with TBI may experience in their daily life. Both Russell et al. and Arenth et al. used a mixture of line drawings of pictures and shapes, words, and letter strings, and only short-term retention was assessed 20,21 . Gillis and Hampstead addressed these methodological shortcomings by using realistic images of common objects and assessed long-term retention outside of the scanner 22 . However, participants recruited into their study were in the chronic stages of recover (range 1 to 13 years). Characterisation of memory for ecologically relevant episodic stimuli following TBI is lacking for the early period of recovery. Here, we rectify this gap by investigating the extent to which memory for common stimuli is impaired following TBI, focussing specifically on the temporal lobe. Processing of common stimuli such as faces, scenes, and animals has been found to recruit specialised temporal lobe structures 23,24 . Exposure to faces robustly activates regions in the middle fusiform gyrus (‘fusiform face area’), the lateral inferior occipital gyrus (‘occipital face area’), and right superior temporal sulcus 25-27 ; exposure to scenes activates the posterior parahippocampus (parahippocampal place area) 28 ; and exposure to animals recruits activity in the bilateral fusiform gyrus 23,29 . Although some degree of impairment is expected given the high prevalence of temporal pathology and specialised processing of various stimuli in this area, it is possible that not all stimuli are equally impaired following injury. Past studies of amnestic patients have demonstrated stimulus-sensitive impairments for complex stimuli such as faces and scenes 30,31 . In the TBI population, impairment of face recognition has also been previously documented 32 . The temporal lobes play a critical role in episodic memory, which is frequently impaired following TBI. Therefore, the present study specifically focusses on the temporal lobes to examine how episodic memory impairments may affect aspects of everyday memory in the early phase of recovery. We used an fMRI task to measure temporal lobe processing during encoding of three stimuli of varying complexity: faces, scenes, and animals. Recognition memory recognition was subsequently probed in an out-of-scanner behavioural task. In line with previous studies 20-22 , we hypothesised that the TBI group would display greater temporal lobe activation during stimulus encoding compared to healthy controls. Furthermore, we hypothesised that greater activity during encoding would be associated with poorer recognition task performance. Lastly, we hypothesised that individuals with TBI would be most impaired for complex stimuli such as faces and scenes. Materials And Methods Participants Twenty-five patients (17 males, 8 females) who had sustained moderate to severe TBI, determined prospectively using the Westmead Post Traumatic Amnesia Scale 33 , were recruited from the Acquired Brain Injury Ward at Epworth Hospital (Richmond, Victoria) after emerging from post-traumatic amnesia ( M = 2.16 months, SD = 1.48 months, range = 0.69 – 6.64 months; Table 1 and Supplementary Table S1). TBI patients predominantly had prefrontal and temporal pathology (Fig. 1). Exclusion criteria included age 75 years, prior history of TBI or other neurological conditions, significant psychiatric or substance abuse history, and MRI contraindication. Twenty-one healthy controls (13 males, 8 females) of similar age, sex, and education were also recruited (Table 1). There were no significant group differences on any of the demographic variables ( P > 0.05). Due to a technical error during data collection, a subset of the sample (6 TBIs and 8 healthy controls) was excluded from the MRI analyses due to poor coverage of the temporal lobes. Additionally, 1 healthy control participant was excluded due to excessive movement in the scanner. Despite this, it should be noted that our fMRI sample size is still comparable to previous neuroimaging studies 20-22 of episodic memory in the TBI population. Written informed consent was provided by all participants. This study was approved by Monash Health Human Research Ethics Committee and conducted in accordance to the Declaration of Helsinki. Episodic memory paradigm The episodic memory paradigm is a task adapted from Mundy et al. to probe episodic memory encoding and recognition (Fig. 2) 31,34 . Participants were presented images of faces, scenes, and animals while in the scanner. They were instructed to respond, using trigger buttons, to each stimulus based on set criteria to ensure attention was maintained throughout the task (e.g. decide whether an animal is shorter or taller than a human man; whether the face is male or female; whether a scene looks hot or cold). consisted of six blocks, each block comprising 20 images from the three stimulus categories. This task comprised 60 unique stimuli, each presented twice throughout the session. Each stimulus was presented for 3 seconds, followed by a 3 second inter-stimulus interval. Five rest blocks were presented after the 1 st – 5 th experimental blocks. Memory of the stimuli presented in the fMRI task was probed in an out of scanner task. Recognition of the episodic stimuli was assessed by instructing participants to classify stimuli as “old” (i.e. images seen during the fMRI task) or “new” (i.e. images that were not presented during the fMRI task). Participants rated a total of 60 old and 60 new stimuli, each presented twice (240 stimuli in total). To assess retrieval ‘confidence’, images were rated on a scale from “definitely old” to “definitely new”. Responses were considered correct/incorrect regardless of confidence level (secondary analysis indicated no significant differences in confidence rating between groups – see “Confidence rating analysis” in Supplementary for further detail). Performance for this task was assessed using accuracy and reaction times. MRI acquisition Structural and functional MR images were acquired in two clinical scanners, using 3.0 Tesla Siemens Magnetom Skyra scanners and a 20-channel head coils. Functional images were acquired using single-shot gradient-echo planar imaging (EPI) with the following parameters: repetition time (TR) = 2.75 s; echo time (TE) = 30 ms; flip angle = 90°; 220 × 220 matrix; voxel size = 3.4 × 3.4 × 3.0 mm. A high-resolution 3D T1-weighted image covering the entire brain was also acquired for anatomical reference (TR = 2.3 s; TE = 2.32 ms; flip angle = 8°; 236 × 350 matrix; voxel size = 0.9 × 0.9 × 0.9 mm). Due to reduced brain coverage (due to using clinical scanners), and a-priori hypotheses, we focused on the temporal lobes. Statistical analysis Behavioural and demographic data Behavioural and demographic data were analysed using R version 3.6.0 (R Core Team, 2019). Two-tailed independent samples t-tests were used to examine group differences on the demographic variables including age, sex, and years of education. Behavioural data were screened for normality, transformed (if necessary), and assessed for violation of statistical assumptions prior to analysis. Outcome measures were analysed using linear mixed models to account for clustering or non-independence of measures within participants. Memory performance was assessed using dprime and reaction time. Dprime measured task accuracy, accounting for the signal (hits) and noise (false alarms). Reaction times were generated by obtaining the average reaction time per stimulus category. Task accuracy was assessed by modelling stimulus category, group, and their interaction (stimulus category x group) as fixed effects, and participant as a random effect. For reaction time, the data were inversely transformed, and a model was fitted with stimulus category, group, and an interaction (stimulus category x group) as fixed effects, and participant as a random effect. Age and education were also added as covariates in the models, given their influence on memory performance 35,36 and reaction time 37 . Where appropriate, post-hoc analyses were conducted using two-tailed t-tests multiple comparison correction. Imaging data MRI preprocessing Prior to preprocessing, lesions were manually segmented using MRIcron ( http://www.mricro.com/mricron ). Preprocessing was performed using fMRIPrep 20.0.0 (Esteban et al., 2018) and involved the application of the following step: undistortion of EPI data, realignment, normalisation, and estimation of confounds. Further information about the MRI preprocessing can be accessed in the Supplementary (see “Detailed MRI preprocessing”). fMRI analysis fMRI data were analysed using FSL’s FEAT version 6.0.2 (FMRIB's Software Library, www.fmrib.ox.ac.uk/fsl). In the first level analysis, contrasts between each stimulus category (i.e. animals, faces, scenes) and the rest blocks were generated for each participant. The onset times for each contrast corresponded to the first stimulus presentation of each category and were 27 seconds in duration. To reduce motion-related artifacts, additional regressors using a modified method of the anatomical CompCor that explained 50% of the variance were also included in the first level model 38 . Differences in brain activation of categories of episodic stimuli were assessed in a 2 (group: TBI vs. healthy controls) x 3 (stimulus category: faces, scenes, and animals) factorial design using FLAME 1+2 mixed effects with automatic outlier de-weighting. An additional explanatory variable was also added at the group level to control for the acquisition of images from two scanners. Given our hypothesis regarding structures in the temporal lobes and their role in learning and processing of category-specific stimuli, an a-priori region of interest (ROI) mask of the temporal lobe was generated using the MNI Structural Atlas (Supplementary Fig. S1) and used in the group level analysis. Imaging findings are reported using a cluster level threshold of Z > 3.1 and a family wise error cluster correction threshold of P < 0.05. FEAT contrast of parameter estimates (COPE) were extracted from significant clusters at the group level. To investigate differences in BOLD response, two-tailed independent samples t-tests were conducted using COPE values. Finally, association between BOLD response and behavioural performance on the episodic recognition task were assessed using Pearson correlations. Results Episodic memory is selectively impaired for face stimuli First, we investigated whether recognition of episodic memories was impaired following TBI. To do this, we performed a linear mixed model assessing memory recall accuracy (old vs. new, as measured using dprime) on the episodic retrieval task. Overall, the TBI group demonstrated significantly poorer recognition accuracy than healthy controls (95% CI, 0.04 – 0.72; P = 0.028; Fig. 3). Post-hoc analyses indicated that the group difference was driven by a significant difference in accuracy for faces (95% CI, 1.11 – 1.50; P = 0.037). There was a trend for lower accuracy for scenes, however, this did not reach statistical significance (95% CI, 0.94 – 1. 19; P = 0.169). There was no significant difference between the groups in accuracy for animals (95% CI, 1. 12 – 1. 26; P = 0.440). Reaction times were quickest for face stimuli Next, we investigated whether reaction time was significantly different between the groups and whether it varied according to the type of stimulus. Overall, reaction time was greater for individuals with TBI compared to healthy controls, irrespective of stimulus type (95% CI, 0.08 – 0.25; P < 0.001; Fig. 3). Post-hoc analyses revealed that the TBI group was significantly slower than healthy controls in responding to faces (95% CI, 0.48 – 0.64; P < 0.001), scenes (95% CI, 0.43 – 0.58; P = 0.001), and animals (95% CI, 0.42 – 0.58; P = 0.001). Across both groups, reaction time also varied depending on the stimulus category: the TBI group was quicker to respond to faces than scenes (95% CI, 0.43 – 0.48; P = 0.014) and animals (95% CI, 0.42 – 0.48; P = 0.005); similarly, healthy controls were quicker to respond to faces than scenes (95% CI, 0.58 – 0.64; P = 0.003) and animals (95% CI, 0.58 – 0.64; P = 0.001). Given this pattern of results, we further investigated whether the poorer performance for faces in the TBI group was driven by a speed-accuracy trade-off. To do this, we included reaction time as a covariate in the linear mixed model. Results indicated that participants with TBI still performed significantly poorer than healthy controls after controlling for reaction time (95% CI, 1. 10 – 1. 54; P = 0.027), suggesting that this pattern of performance was not due solely to a speed-accuracy trade-off. fMRI Task activates the stereotypical regions underpinning encoding of episodic stimuli To demonstrate that our task elicited activations in stereotypical areas involved with the canonical network that support encoding of episodic stimuli, we first included all participants in an analysis looking at the average activation for each stimulus category (i.e. faces, scenes, and animals; Fig. 4). During encoding of face stimuli, significant clusters were noted in face-selective areas including the right inferior occipital gyrus and left/right fusiform gyrus 25-27 , as well as the right hippocampus. During encoding of scene stimuli, significant clusters were noted in scene-selective area of right parahippocampal gyrus 28 , as well as the left fusiform gyrus. Finally, during encoding of animal stimuli, significant clusters were noted in animal-selective area of the right fusiform gyrus 23,29 , as well as the left inferior occipital gyrus. TBI patients show reduced right transverse temporal gyrus activation during face processing Consistent with the behavioural results, group differences on imaging were apparent during encoding of faces. TBI participants showed reduced activation in the right transverse temporal gyrus extending to the planum temporale compared to healthy controls (Fig. 5). FEAT analysis COPE values were extracted for the significant cluster. We first investigated whether there was a significant difference in BOLD response between groups using an independent samples t-test. As expected, TBI patients displayed lower COPE values ( M = -9.90, SD = 24.79) compared to healthy controls ( M = 19.07, SD = 33.04), t (18) = -2.61, P = 0.017. We further examined whether there was an association with behavioural performance on the episodic recognition task using Pearson correlations. Overall, there was a moderate positive relationship between COPE values and the dprime scores for face stimuli, r (28) = 0.497, P = 0.005. Follow-up correlations indicated that a significant correlation was only apparent for the TBI group, r (16) = 0.480, P = 0.044, and not healthy controls, r (10) = 0.375, P = 0.229. To examine specificity of this brain-behaviour relationship, we conducted a control analysis, correlating whole temporal lobe activity with dprime scores for face stimuli. Whole temporal lobe activation was not correlated with face recognition performance ( P > 0.05; see “Additional control analyses” in Supplementary). TBI patients display reduced right fusiform gyrus activity during scene processing Despite a non-significant difference between groups for scene recognition, we found reduced activation in the right posterior fusiform gyrus for the TBI group in comparison to healthy controls during scene encoding (Fig. 6). TBI patients showed lower COPE values ( M = 29.26, SD = 25.11) compared to controls ( M = 76.93, SD = 47.76), t (15) = 3.18, P = 0.006. To further examine whether functional changes in this cluster were associated with behavioural performance, we again conducted a series of Pearson correlations. Upon removal of an outlier (see “Additional scene cluster results” in Supplementary for further detail), there was a moderate postive relationship between the COPE values extracted from this cluster and the dprime scores for scene stimuli, r (27) = 0.482, P = 0.008. Further examation revealed that a significant correlation was only apparent for the TBI group, r (15) = 0.574, P = 0.016, and not healthy controls, r (10) = 0.317, P = 0.316. Similar to the control analysis above, we conducted a control analysis, correlating whole temporal lobe activity with dprime scores for scene stimuli. Interestingly, general temporal lobe activation was correlated with scene recognition performance ( P = 0.006). However, correlations were no longer significant once the groups were examined separately ( P > 0.05; see “Additional control analyses” in Supplementary). Discussion The present study focused on determining the role of temporal lobe activity in episodic memory behaviour following TBI. We showed for the first time, using converging evidence from behavioural and fMRI data, that episodic memory impairment following TBI appeared to be category-specific and related specific sub-regions within the temporal lobes. Deficits were most apparent for faces; TBI patients displayed reduced transverse temporal gyrus activation during face encoding and subsequent impairment on face recognition. TBI patients also displayed reduced fusiform gyrus activation for scenes; this is despite no statistically significant difference between groups during scene recognition. Interestingly, brain activation during face and scene encoding correlated with subsequent recognition for TBI patients but not in healthy control participants. Overall, these findings suggest that TBI: a) preferentially impairs episodic memory in specific domains, and b) aberrant neural processing may not be reflected in statistically significant differences behavioural assessments, and thus neural activation conveys complimentary information undetected through examination of overt behaviour. Broadly speaking, our findings are similar to those of previous studies of amnestic patients who show stimulus-sensitive impairments for complex stimuli such as faces and scenes 30,31 . Our findings are also in concordance with a previous study demonstrating impaired face recognition in the TBI population 32 . Valentine et al. subjected participants to a range of facial recognition and learning tasks and found that while performance varied, deficits were more apparent for tasks with greater demands 32 . More specifically, the most sensitive tasks were those which contained a larger number of faces to be encoded or had fewer presentations of the stimuli. Our task was comparably difficult in that participants were presented with a similar number of face stimuli which were only shown twice during the encoding phase; thus, it was not surprising we obtained a similar finding. As expected, we found that the TBI group was generally slower than healthy controls in their reaction times. Both groups, however, responded more quickly to faces than to animals and scenes. This result somewhat aligns with a study conducted by Keightley et al. who found that participants reacted quicker to faces than scenes, despite accuracy being better for scene stimuli 39 . We further explored whether speed-accuracy trade off could account for our findings, given participants displayed the poorest accuracy for face stimuli. We found that including reaction time as a covariate when determining between-group differences in face accuracy did not change the result. Instead, the rapid response to faces suggests that individuals with TBI may have performed superficial encoding of face stimuli, thus negatively impacting decision-making during recognition. A unique aspect of our study was the inclusion of animal and scene stimuli, in addition to faces, which allowed us to investigate the effects of stimulus complexity on episodic memory. Interestingly, we found the TBI group displayed altered neural activity without showing statistically significant impairment of behavioural performance for scene stimuli, whereas changes on both functional and behavioural measures were apparent for faces. One potential reason for these results is that faces are more visually complex than scenes. Indeed, facial processing is a complex phenomenon requiring multifaceted processes across widespread cortical areas 40 . Unlike most other visual stimuli they are processed in a holistic and configural manner 41,42 ; thus, discrimination requires attention to detail and subtle perception of variable facial features 40 . Our fMRI results provide further insights into the mechanisms that may underlie behavioural deficits for faces. We found that the TBI group showed a reduced response in the right transverse temporal gyrus extending to the planum temporale during encoding of faces. However, these areas do not form part of the core or extended face network 26 . The transverse temporal gyrus is predominantly implicated in auditory processing 43,44 , although studies have also demonstrated its role in spontaneous inner speech 45 . The planum temporale has been shown to be involved in language functions 46 . Therefore, it is possible that difference in BOLD activity in this temporal lobe region may reflect differences in verbalisation during face encoding. Interestingly, activation of this region correlated with behavioural performance only for individuals with TBI. It may be that individuals with TBI may therefore have made less attempts to verbally describe the stimuli when processing faces. This further supports the hypothesis that TBI participants encoded these stimuli in a rapid and superficial manner. In contrast, healthy controls did not display such an association, suggesting verbalisation during face encoding was not key in determining later recognition success. Our other key finding was that the TBI group demonstrated reduced brain activity during encoding of scenes in the right posterior fusiform gyrus. The right posterior fusiform gyrus responds non-selectively to faces and scenes and generally may be involved with processing complex visual stimuli 47 . This result suggests impaired recruitment of this temporal lobe sub-region during processing of scenes. Indeed, we found that individuals with TBI who had higher activation within this region performed better during scene recognition. Despite differences in brain activation, the groups did not differ with respect to behaviour. This finding may be due to scene stimuli containing a greater number of contextual cues that could further aid encoding and recollection. For example, individuals may have used cues such as the location (e.g. kitchen) or remembered certain salient scene features (e.g. item/s contained in the scene). To examine the specificity of the brain-behaviour relationships, we conducted a series of control analyses to determine whether the association between brain activation and behavioural performance was specific to the significant clusters or, more generally, to activity in the temporal lobes. In support of our main findings, general activation in the temporal lobes was not associated with face recognition accuracy, thus highlighting the specificity of right transverse temporal gyrus in predicting face recognition performance. However, we found that general activation in the temporal lobes was associated scene recognition accuracy. This relationship disappeared when the TBI group and healthy controls were considered separately. This indicates that the right fusiform gyrus has less specificity in predicting scene recognition performance. In line with our behaviour findings, these results suggest greater, more robust, specificity for face stimuli, compared to scenes. Although our findings show a clear brain-behaviour relationship, these associations oppose our initial hypotheses, which predicted that individuals with TBI would display greater temporal lobe activity in support of previous studies 20-22 . One general model that could explain this finding is that of cortical reorganisation following injury 48-50 . That is, the pattern of cortical activation reflecting neural compensation or recovery following TBI is likely to depend on the length of time since an individual’s injury 51 . A key methodological difference is that our TBI participants were recruited at an average of 2 months’ post-injury whereas those in past studies were recruited over 1 year post-injury{Arenth, 2012 #40} 20-22 {Arenth, 2012 #40}{Russell, 2011 #14}{Arenth, 2012 #40}{Arenth, 2014 #520}. In a key study, Sanchez-Carrion et al. characterised the longitudinal changes in brain activity following TBI 52 . They did so by assessing brain activity during a working memory task at 6 months and 1 year following the injury. Their findings show an initial reduction in brain activity at the 6 month time-points which gradually resolved by 1 year following the injury. Thus, discrepancies in our findings from past studies may reflect differences in recovery phases. Our study has several important implications. From a clinical perspective, it is generally acknowledged that individuals have generalised episodic memory deficits after injury. Our findings provide evidence to the contrary and show that impairment is more apparent with complex visual stimuli such as faces and scenes. An obvious clinical translation is the need to provide strategies that promote deeper processing to better aid memory for these and other complex stimuli. In addition, our study provides further support for the utility of fMRI as a complimentary source of information by demonstrating evidence of aberrant neural processing that was less evident in behavioural performance. This is important considering that most assessments of memory are based on behavioural performance. There were some limitations in our study, however. Our episodic memory paradigm only allowed us to investigate functional activity during stimulus encoding. Therefore, we could not comment on the extent in which temporal structures are implicated in recognition of episodic stimuli following TBI. This may be an avenue for exploration in future research. Although not a limitation per se, we took an a-priori approach specifically focusing on the temporal lobes. While this allowed us to answer specific questions about the temporal lobes’ role in processing of episodic stimuli, it also limited our investigation of non-temporal contributions to episodic memory. For example, there is value in investigating the interaction between frontal and temporal regions to support encoding of more complex stimuli, especially considering the frontal lobes’ role in strategy, allocation of resources, and planning 53,54 . In conclusion, we found evidence demonstrating that individuals with TBI show impairment of episodic memory for complex stimuli and that this was associated with functional changes. In comparison to healthy controls, we found that the TBI group displayed reduced activation in the right transverse temporal gyrus and fusiform gyrus during face and scene processing, respectively. We found that brain activation in these temporal lobe sub-regions were associated with behavioural performance for the TBI group and not healthy controls. These findings may be explained in terms of differences in verbalisation during encoding and cortical reorganisation. Overall, we provide preliminary evidence demonstrating that following TBI: a) episodic memory impairment is domain specific and more broadly dependent on the complexity of the stimuli, and b) aberrant neuronal activity may exist despite lack of evidence of significant impairment in behavioural performance, and therefore neural activation may be a more robust early indicator than behaviour. Abbreviations TBI = traumatic brain injury; functional magnetic resonance imaging = fMRI; GSC = Glasgow Coma Scale; Post Traumatic Amnesia = PTA; WPTAS = Westmead Post Traumatic Amnesia Scale; DAI = diffuse axonal injury; EDH = extradural haematoma; ICH = intracerebral haemorrhage; SAH = subarachnoid haemorrhage; SDH = subdural haemorrhage; NAD = no abnormality detected; TR = repetition time; TE = echo time; contrast of parameter estimates = COPE Declarations Acknowledgements GS was funded by a National Health and Research Council Early Career Fellowship (APP1104692) and the Brain Foundation. This work was supported by the Multi-modal Australian ScienceS Imaging and Visualisation Environment (MASSIVE) HPC facility ( www.massive.org.au ). In addition, the authors would like to thank the staff at the Acquired Brain Injury Ward at Epworth Hospital (Richmond) and Bridge Road Imaging. The authors would also like to thank the participants who took part in the study. Authors contribution AT collected/analysed the data and wrote the manuscript; MM conceptualised the study, designed the study, and reviewed the manuscript; JP conceptualised the study and reviewed the manuscript; GS conceptualised the study, designed the study, collected/analysed the data, and reviewed the manuscript. Ethics declaration Completing interests The authors declare no competing interests. Data availability All data supporting the findings of this study can be requested from the corresponding author. References 1 Bigler, E. D. The lesion (s) in traumatic brain injury: Implications for clinical neuropsychology. Archives of clinical neuropsychology 16 , 95-131 (2001). 2 Draper, K. & Ponsford, J. Cognitive functioning ten years following traumatic brain injury and rehabilitation. Neuropsychology 22 , 618-625, doi:10.1037/0894-4105.22.5.618 (2008). 3 Azouvi, P., Arnould, A., Dromer, E. & Vallat-Azouvi, C. Neuropsychology of traumatic brain injury: An expert overview. Rev Neurol (Paris) 173 , 461-472, doi:10.1016/j.neurol.2017.07.006 (2017). 4 Tittle, A. & Burgess, G. H. Relative contribution of attention and memory toward disorientation or post-traumatic amnesia in an acute brain injury sample. Brain Injury 25 , 933-942, doi:10.3109/02699052.2011.597042 (2011). 5 Rabinowitz, A. R. & Levin, H. S. Cognitive sequelae of traumatic brain injury. Psychiatr Clin North Am 37 , 1-11, doi:10.1016/j.psc.2013.11.004 (2014). 6 Vakil, E. The effect of moderate to severe traumatic brain injury (TBI) on different aspects of memory: a selective review. J Clin Exp Neuropsychol 27 , 977-1021, doi:10.1080/13803390490919245 (2005). 7 Moscovitch, M., Cabeza, R., Winocur, G. & Nadel, L. Episodic Memory and Beyond: The Hippocampus and Neocortex in Transformation. Annu Rev Psychol 67 , 105-134, doi:10.1146/annurev-psych-113011-143733 (2016). 8 Tulving, E. Episodic memory: from mind to brain. Annual review of psychology 53 , 1-25 (2002). 9 Nakase-Richardson, R. et al. Utility of post-traumatic amnesia in predicting 1-year productivity following traumatic brain injury: comparison of the Russell and Mississippi PTA classification intervals. J Neurol Neurosurg Psychiatry 82 , 494-499, doi:10.1136/jnnp.2010.222489 (2011). 10 Wagner, A. D., Shannon, B. J., Kahn, I. & Buckner, R. L. Parietal lobe contributions to episodic memory retrieval. Trends Cogn Sci 9 , 445-453, doi:10.1016/j.tics.2005.07.001 (2005). 11 Eichenbaum, H. Prefrontal-hippocampal interactions in episodic memory. Nat Rev Neurosci 18 , 547-558, doi:10.1038/nrn.2017.74 (2017). 12 Dickerson, B. C. & Eichenbaum, H. The episodic memory system: neurocircuitry and disorders. Neuropsychopharmacology 35 , 86-104, doi:10.1038/npp.2009.126 (2010). 13 Simons, J. S. & Spiers, H. J. Prefrontal and medial temporal lobe interactions in long-term memory. Nat Rev Neurosci 4 , 637-648, doi:10.1038/nrn1178 (2003). 14 Graham, K. S., Barense, M. D. & Lee, A. C. Going beyond LTM in the MTL: a synthesis of neuropsychological and neuroimaging findings on the role of the medial temporal lobe in memory and perception. Neuropsychologia 48 , 831-853, doi:10.1016/j.neuropsychologia.2010.01.001 (2010). 15 Cameron, K. A., Yashar, S., Wilson, C. L. & Fried, I. Human hippocampal neurons predict how well word pairs will be remembered. Neuron 30 , 289-298 (2001). 16 Barlow, K. M. Traumatic brain injury. Handb Clin Neurol 112 , 891-904, doi:10.1016/B978-0-444-52910-7.00011-8 (2013). 17 Daneshvar, D. H. & McKee, A. C. Traumatic Brain Injury. 219-235, doi:10.1016/b978-0-12-398270-4.00016-1 (2015). 18 Bigler, E. D. et al. Traumatic brain injury and memory: The role of hippocampal atrophy. Neuropsychology 10 , 333 (1996). 19 Ariza, M. et al. Hippocampal head atrophy after traumatic brain injury. Neuropsychologia 44 , 1956-1961, doi:10.1016/j.neuropsychologia.2005.11.007 (2006). 20 Russell, K. C., Arenth, P. M., Scanlon, J. M., Kessler, L. J. & Ricker, J. H. A functional magnetic resonance imaging investigation of episodic memory after traumatic brain injury. J Clin Exp Neuropsychol 33 , 538-547, doi:10.1080/13803395.2010.537253 (2011). 21 Arenth, P. M., Russell, K. C., Scanlon, J. M., Kessler, L. J. & Ricker, J. H. Encoding and recognition after traumatic brain injury: neuropsychological and functional magnetic resonance imaging findings. J Clin Exp Neuropsychol 34 , 333-344, doi:10.1080/13803395.2011.633896 (2012). 22 Gillis, M. M. & Hampstead, B. M. A two-part preliminary investigation of encoding-related activation changes after moderate to severe traumatic brain injury: hyperactivation, repetition suppression, and the role of the prefrontal cortex. Brain Imaging Behav 9 , 801-820, doi:10.1007/s11682-014-9337-5 (2015). 23 Downing, P. E., Chan, A. W., Peelen, M. V., Dodds, C. M. & Kanwisher, N. Domain specificity in visual cortex. Cereb Cortex 16 , 1453-1461, doi:10.1093/cercor/bhj086 (2006). 24 Mundy, M. E. et al. Material-independent and material-specific activation in functional MRI after perceptual learning. Neuroreport 20 , 1397-1401 (2009). 25 Hoffman, E. A. & Haxby, J. V. Distinct representations of eye gaze and identity in the distributed human neural system for face perception. Nature neuroscience 3 , 80-84 (2000). 26 Haxby, J. V., Hoffman, E. A. & Gobbini, M. I. Human neural systems for face recognition and social communication. Biological psychiatry 51 , 59-67 (2002). 27 Kesler, M. L. et al. Neural substrates of facial emotion processing using fMRI. Cognitive Brain Research 11 , 213-226 (2001). 28 Epstein, R. A. & Ward, E. J. How reliable are visual context effects in the parahippocampal place area? Cereb Cortex 20 , 294-303, doi:10.1093/cercor/bhp099 (2010). 29 Rogers, T. T., Hocking, J., Mechelli, A., Patterson, K. & Price, C. Fusiform activation to animals is driven by the process, not the stimulus. Journal of Cognitive Neuroscience 17 , 434-445 (2005). 30 Taylor, K. J., Henson, R. N. & Graham, K. S. Recognition memory for faces and scenes in amnesia: dissociable roles of medial temporal lobe structures. Neuropsychologia 45 , 2428-2438, doi:10.1016/j.neuropsychologia.2007.04.004 (2007). 31 Mundy, M. E., Downing, P. E., Dwyer, D. M., Honey, R. C. & Graham, K. S. A critical role for the hippocampus and perirhinal cortex in perceptual learning of scenes and faces: complementary findings from amnesia and FMRI. Journal of Neuroscience 33 , 10490-10502 (2013). 32 Valentine, T., Powell, J., Davidoff, J., Letson, S. & Greenwood, R. Prevalence and correlates of face recognition impairments after acquired brain injury. Neuropsychol Rehabil 16 , 272-297, doi:10.1080/09602010500176443 (2006). 33 Shores, E. A., Marosszeky, J., Sandanam, J. & Batchelor, J. Preliminary validation of a clinical scale for measuring the duration of post-traumatic amnesia. Med J Aust 144 , 569-572 (1986). 34 Mundy, M. E., Downing, P. E. & Graham, K. S. Extrastriate cortex and medial temporal lobe regions respond differentially to visual feature overlap within preferred stimulus category. Neuropsychologia 50 , 3053-3061 (2012). 35 Hoyer, W. J. & Verhaeghen, P. in Handbook of the psychology of aging 209-232 (Elsevier, 2006). 36 Lachman, M. E., Agrigoroaei, S., Murphy, C. & Tun, P. A. Frequent cognitive activity compensates for education differences in episodic memory. The American Journal of Geriatric Psychiatry 18 , 4-10 (2010). 37 Der, G. & Deary, I. J. The relationship between intelligence and reaction time varies with age: Results from three representative narrow-age age cohorts at 30, 50 and 69 years. Intelligence 64 , 89-97 (2017). 38 Muschelli, J. et al. Reduction of motion-related artifacts in resting state fMRI using aCompCor. Neuroimage 96 , 22-35, doi:10.1016/j.neuroimage.2014.03.028 (2014). 39 Keightley, M. L., Chiew, K. S., Anderson, J. A. & Grady, C. L. Neural correlates of recognition memory for emotional faces and scenes. Soc Cogn Affect Neurosci 6 , 24-37, doi:10.1093/scan/nsq003 (2011). 40 Haxby, J. V., Hoffman, E. A. & Gobbini, M. I. The distributed human neural system for face perception. Trends Cogn. Sci. 4 , 223-233 (2000). 41 Park, J., Newman, L. I. & Polk, T. A. Face processing: the interplay of nature and nurture. The Neuroscientist 15 , 445-449 (2009). 42 Maurer, D., Le Grand, R. & Mondloch, C. J. The many faces of configural processing. Trends Cogn. Sci. 6 , 255-260 (2002). 43 Kaas, J. H., Hackett, T. A. & Tramo, M. J. Auditory processing in primate cerebral cortex. Curr. Opin. Neurobiol. 9 , 164-170 (1999). 44 Warrier, C. et al. Relating structure to function: Heschl's gyrus and acoustic processing. J. Neurosci. 29 , 61-69 (2009). 45 Hurlburt, R. T., Alderson-Day, B., Kühn, S. & Fernyhough, C. Exploring the ecological validity of thinking on demand: neural correlates of elicited vs. spontaneously occurring inner speech. PloS one 11 , e0147932 (2016). 46 Shapleske, J., Rossell, S. L., Woodruff, P. & David, A. The planum temporale: a systematic, quantitative review of its structural, functional and clinical significance. Brain Research Reviews 29 , 26-49 (1999). 47 Nakamura, K. et al. Functional delineation of the human occipito-temporal areas related to face and scene processing: a PET study. Brain 123 , 1903-1912 (2000). 48 Levine, B. et al. Functional reorganisation of memory after traumatic brain injury: a study with H2150 positron emission tomography. Journal of Neurology, Neurosurgery & Psychiatry 73 , 173-181 (2002). 49 Christodoulou, C. et al. Functional magnetic resonance imaging of working memory impairment after traumatic brain injury. Journal of Neurology, Neurosurgery & Psychiatry 71 , 161-168 (2001). 50 Hillary, F. G. Neuroimaging of working memory dysfunction and the dilemma with brain reorganization hypotheses. J Int Neuropsychol Soc 14 , 526-534, doi:10.1017/S1355617708080788 (2008). 51 Munoz-Cespedes, J. M., Rios-Lago, M., Paul, N. & Maestu, F. Functional neuroimaging studies of cognitive recovery after acquired brain damage in adults. Neuropsychol Rev 15 , 169-183, doi:10.1007/s11065-005-9178-5 (2005). 52 Sanchez-Carrion, R. et al. A longitudinal fMRI study of working memory in severe TBI patients with diffuse axonal injury. Neuroimage 43 , 421-429, doi:10.1016/j.neuroimage.2008.08.003 (2008). 53 Vakil, E., Greenstein, Y., Weiss, I. & Shtein, S. The Effects of Moderate-to-Severe Traumatic Brain Injury on Episodic Memory: a Meta-Analysis. Neuropsychol Rev 29 , 270-287, doi:10.1007/s11065-019-09413-8 (2019). 54 Stuss, D. T. & Alexander, M. P. Does damage to the frontal lobes produce impairment in memory? Current Directions in Psychological Science 14 , 84-88 (2005). Tables Table 1. Demographic information and clinical characteristics of the TBI and healthy groups. Demographic variables Traumatic brain injury Mean (SD) Healthy controls, Mean (SD) Age (years) 38.36 (16.82) 37.29 (17.89) Sex (male/female) 17/8 13/8 Education (years) 13.82 (3.27) 14.19 (2.57) Time since injury (months) 2.16 (1.51) - PTA (days) 22.46 (14.63) - GCS (lowest) 9 (4.37) - GSC = Glasgow Coma Scale; PTA = post-traumatic amnesia. Note: PTA duration were available for n = 24 patients. Additional Declarations No competing interests reported. Supplementary Files EpisodicpaperrevisedsupplementaryScientificReports.docx Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Major revision 21 Apr, 2021 Reviews received at journal 19 Mar, 2021 Reviewers agreed at journal 15 Mar, 2021 Reviewers invited by journal 14 Mar, 2021 Editor assigned by journal 11 Mar, 2021 Editor invited by journal 28 Jan, 2021 Submission checks completed at journal 25 Jan, 2021 First submitted to journal 21 Jan, 2021 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-152808","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":8974602,"identity":"34bde02e-683c-4ddf-8da3-71377cd5a621","order_by":0,"name":"Abbie Taing","email":"","orcid":"","institution":"Monash University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Abbie","middleName":"","lastName":"Taing","suffix":""},{"id":8974603,"identity":"4dc63a21-eb3b-4c7f-9120-3bc2ac3c1f42","order_by":1,"name":"Matthew Mundy","email":"","orcid":"","institution":"Monash University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Matthew","middleName":"","lastName":"Mundy","suffix":""},{"id":8974604,"identity":"18877f2f-2bdb-4752-ad7a-b73904c93b1c","order_by":2,"name":"Jennie Ponsford","email":"","orcid":"","institution":"Monash University","correspondingAuthor":false,"submittingAuthor":false,"prefix":"","firstName":"Jennie","middleName":"","lastName":"Ponsford","suffix":""},{"id":8974605,"identity":"d4704e4a-4e3a-431f-bff6-2158cb320176","order_by":3,"name":"Gershon Spitz","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABCUlEQVRIiWNgGAWjYJACZgjF2MDAUAFmsSEECWhpbGA4Q5oWoDWMbURokXfvffy5oOIOg/zs5vaHP+dtkzNnP/7sAUOFdWIDDi2GZ46bSc8484zB4M7BxgbJbbeNLXtyzA0YzqTj1jIjjY2Zt+0wg4FEYmOD4bbbiRsO5LBJMLYdxq1l/jPmzyAt8jOAWhLnALWcf/5MgvEfbi3yEmwM0iAtDDeAWg42ALXcSDCTYGzArcWAJ41NmufMYR6QX2Y2HLttbHDjjZlEwrF0Y5y2tB9j/sxTcVhOfnb7g48/am7LGZxPfybxocZaFqctByA0D4MEsnACDuVgW+BmSeBRNQpGwSgYBSMbAAAqil6BEk7SaAAAAABJRU5ErkJggg==","orcid":"","institution":"Monash Epworth Rehabilitation Research Centre, 185-187 Hoddle Street, Richmond, Victoria 3121, Australia.","correspondingAuthor":true,"submittingAuthor":false,"prefix":"","firstName":"Gershon","middleName":"","lastName":"Spitz","suffix":""}],"badges":[],"createdAt":"2021-01-21 23:44:05","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-152808/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-152808/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":5349263,"identity":"f874ca00-b75d-45ac-b7a0-b7d3e28f2978","added_by":"auto","created_at":"2021-01-28 19:16:52","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":55123,"visible":true,"origin":"","legend":"Lesion overlay plot of all TBI participants. Maps were overlaid on a T1 template in MNI space.","description":"","filename":"Fig1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-152808/v1/0f6783413fa2959e79b8cc78.jpg"},{"id":5349261,"identity":"54415897-be95-47a8-a87f-f5e5607fbc95","added_by":"auto","created_at":"2021-01-28 19:16:50","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":41057,"visible":true,"origin":"","legend":"Schematic diagram of the episodic encoding task. Participants were presented 5 images from a stimulus category (i.e. faces, scenes, and animals) and were instructed to response to the various stimuli on screen based on set criteria. Each stimulus was presented for 3 s, followed by an inter-stimulus duration of 3s. ","description":"","filename":"Fig2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-152808/v1/3775c4aedd99d1db3f8b02b6.jpg"},{"id":5349277,"identity":"f96811d3-91d1-4f6d-8cbb-0a52e943f7c1","added_by":"auto","created_at":"2021-01-28 19:17:12","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":61678,"visible":true,"origin":"","legend":"Behavioural results for the episodic retrieval task. a) Plots representing accuracy according to stimulus category, as measured using dprime (higher values denote better performance). The TBI group had significantly poorer accuracy than healthy controls when retrieving faces (P = 0.037). There was a trend for lower accuracy for scenes, although this did not reach statistical significance (P = 0.169). There was no significant difference in accuracy for animal stimuli (P = 0.440). b) Plots representing reaction time according to stimulus category (note: reaction time was inversely transformed; higher values denote faster performance). As expected, the TBI group was slower than healthy controls when responding to the various stimuli (P \u003c 0.05). Note: reported P-values were adjusted for multiple comparisons. ","description":"","filename":"Fig3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-152808/v1/f883e48e01059b5efae1a34a.jpg"},{"id":5349267,"identity":"d31e0966-82b6-45e4-b395-3143179b528e","added_by":"auto","created_at":"2021-01-28 19:17:02","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":60129,"visible":true,"origin":"","legend":"Overall functional activity elicited during the episodic encoding task for the whole sample. Significant clusters during encoding of faces (blue), scenes (green), and animals (red). ","description":"","filename":"Fig4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-152808/v1/a45392f00abce2c8b6148022.jpg"},{"id":5349278,"identity":"20eb0d60-4a65-44b2-89f8-e09c9e9053aa","added_by":"auto","created_at":"2021-01-28 19:17:13","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":96664,"visible":true,"origin":"","legend":"Reduced functional activity during encoding of faces was associated with behavioural performance. a) The TBI group demonstrated a reduced response in the right transverse temporal gyrus extending to the planum temporale during encoding of faces compared to healthy controls. There was no overlap between this cluster and lesions overlap (green). Note: cluster was dilated to increase visibility. b) Plot of the COPE values extracted from the significant cluster in the right transverse temporal gyrus extending to the planum temporale. The difference in COPE values between the TBI group (red) and healthy controls (blue) was significant (P = 0.017). c) Overall, there was a significant association between COPE values and the dprime scores for face stimuli (P = 0.005). However, further investigation indicated that a significant association was only apparent for d) TBI group (P = 0.044), and not e) healthy controls (P = 0.229).","description":"","filename":"Fig5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-152808/v1/9a3cd3a3c4e0a2faeac7e52e.jpg"},{"id":5349268,"identity":"bf3fb7b9-d7ff-4f40-852e-27b1df318305","added_by":"auto","created_at":"2021-01-28 19:17:06","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":96374,"visible":true,"origin":"","legend":"Reduced functional activity during encoding of scenes was associated with behavioural performance. a) In comparison to healthy controls, the TBI group demonstrated reduced activity in the right posterior fusiform gyrus encoding of scene stimuli. There was no overlap between this cluster and lesions overlap (green). Note: cluster was dilated using to increase visibility. b) Plot of the COPE values extracted from the significant cluster in the right posterior fusiform gyrus. The difference in COPE values between the TBI group (red) and healthy controls (blue) was significant (P = 0.006). c) Overall, there was a signification association between the COPE values and the dprime for scene stimuli (P = 0.008). However, further investigation indicated that a significant association was only apparent for d) TBI group (P = 0.016), and not e) healthy controls (P = 0.316).","description":"","filename":"Fig6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-152808/v1/e3f4d8a07bdaf279d33668b3.jpg"},{"id":13653062,"identity":"7b5fdf86-675e-4fd1-809e-2c99f226b121","added_by":"auto","created_at":"2021-09-17 09:51:49","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":759905,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-152808/v1/c5d6438c-6f0a-475d-adae-3b2c7be57e47.pdf"},{"id":5349280,"identity":"67a71e51-816f-40a0-bd24-f967bd611e24","added_by":"auto","created_at":"2021-01-28 19:17:16","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":911964,"visible":true,"origin":"","legend":"","description":"","filename":"EpisodicpaperrevisedsupplementaryScientificReports.docx","url":"https://assets-eu.researchsquare.com/files/rs-152808/v1/3e0a3b51dfc3bf7584d85f53.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"\u003cp\u003eTemporal Lobe Activation Predicts Episodic Memory Following Traumatic Brain Injury\u003c/p\u003e","fulltext":[{"header":"Introduction","content":"\u003cp\u003eTraumatic brain injury (TBI) can cause focal and diffuse disruption to multiple brain systems. Pathology is most widely observed in frontal and temporal cortices\u003csup\u003e1\u003c/sup\u003e, and as such impairments in processing speed, attention, executive function, and memory are most common\u003csup\u003e2,3\u003c/sup\u003e. Invariably, all individuals with a moderate or severe TBI will experience an initial transient period of impaired consciousness with amnesia and general confusion known as post-traumatic amnesia (PTA)\u003csup\u003e4\u003c/sup\u003e. Although most individuals emerge from PTA, many report ongoing difficulty with so-called episodic memories\u003csup\u003e5,6\u003c/sup\u003e. Episodic memories involve the ability to learn, store, and retrieve information about personal experiences\u003csup\u003e7,8\u003c/sup\u003e. Impairment of episodic memory can interfere with crucial skills such as new learning and task completion, and therefore can significantly limit functional independence and productivity\u003csup\u003e9\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eNeuroanatomically, episodic memory is supported by a brain network spanning several regions in the frontal and parietal cortices\u003csup\u003e10,11\u003c/sup\u003e. However, structures contained within the temporal lobes are arguably the most crucial for particular aspects of episodic memory\u003csup\u003e7,12\u003c/sup\u003e. The temporal lobes house the hippocampus, and the entorhinal, perirhinal, and parahippocampal cortices \u0026ndash; collectively known as the medial temporal lobes\u003csup\u003e13,14\u003c/sup\u003e. The hippocampus and other medial temporal structures are involved in the encoding of episodic stimuli, and modulation of activity in these regions has been found to predict subsequent recollection of the encoded stimuli\u003csup\u003e15\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eThe temporal lobes are preferentially affected after TBI\u003csup\u003e1\u003c/sup\u003e. The location of the temporal lobes and their proximity to the middle cranial fossa makes them highly susceptible to acceleration-deceleration forces present during injury\u003csup\u003e16,17\u003c/sup\u003e. Structures within the temporal lobes, such as the hippocampus and other limbic structures, are particularly susceptible to hypoxic insults and excitotoxicity effects\u003csup\u003e12\u003c/sup\u003e. As such, atrophy in the hippocampus and fornix is common following TBI\u003csup\u003e18\u003c/sup\u003e, and has been associated with impaired memory\u003csup\u003e19\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eFunctional imaging studies in the TBI population have also implicated the temporal lobes in impaired episodic memory. Individuals with TBI tend to display increased activity in various regions, including the temporal lobes, during encoding of episodic stimuli when compared with healthy controls\u003csup\u003e20-22\u003c/sup\u003e. One limitation of these previous studies is that most have failed to use paradigms that may reflect the type of episodic memory deficits that individuals with TBI may experience in their daily life. Both Russell et al. and Arenth et al. used a mixture of line drawings of pictures and shapes, words, and letter strings, and only short-term retention was assessed\u003csup\u003e20,21\u003c/sup\u003e. Gillis and Hampstead addressed these methodological shortcomings by using realistic images of common objects and assessed long-term retention outside of the scanner\u003csup\u003e22\u003c/sup\u003e. However, participants recruited into their study were in the chronic stages of recover (range 1 to 13 years). Characterisation of memory for ecologically relevant episodic stimuli following TBI is lacking for the early period of recovery.\u003c/p\u003e\n\u003cp\u003eHere, we rectify this gap by investigating the extent to which memory for common stimuli is impaired following TBI, focussing specifically on the temporal lobe. Processing of common stimuli such as faces, scenes, and animals has been found to recruit specialised temporal lobe structures\u003csup\u003e23,24\u003c/sup\u003e. Exposure to faces robustly activates regions in the middle fusiform gyrus (\u0026lsquo;fusiform face area\u0026rsquo;), the lateral inferior occipital gyrus (\u0026lsquo;occipital face area\u0026rsquo;), and right superior temporal sulcus\u003csup\u003e25-27\u003c/sup\u003e; exposure to scenes activates the posterior parahippocampus (parahippocampal place area)\u003csup\u003e28\u003c/sup\u003e; and exposure to animals recruits activity in the bilateral fusiform gyrus\u003csup\u003e23,29\u003c/sup\u003e. Although some degree of impairment is expected given the high prevalence of temporal pathology and specialised processing of various stimuli in this area, it is possible that not all stimuli are equally impaired following injury. Past studies of amnestic patients have demonstrated stimulus-sensitive impairments for complex stimuli such as faces and scenes\u003csup\u003e30,31\u003c/sup\u003e. In the TBI population, impairment of face recognition has also been previously documented\u003csup\u003e32\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eThe temporal lobes play a critical role in episodic memory, which is frequently impaired following TBI. Therefore, the present study specifically focusses on the temporal lobes to examine how episodic memory impairments may affect aspects of everyday memory in the early phase of recovery. We used an fMRI task to measure temporal lobe processing during encoding of three stimuli of varying complexity: faces, scenes, and animals. Recognition memory recognition was subsequently probed in an out-of-scanner behavioural task. In line with previous studies\u003csup\u003e20-22\u003c/sup\u003e, we hypothesised that the TBI group would display greater temporal lobe activation during stimulus encoding compared to healthy controls. Furthermore, we hypothesised that greater activity during encoding would be associated with poorer recognition task performance. Lastly, we hypothesised that individuals with TBI would be most impaired for complex stimuli such as faces and scenes.\u003c/p\u003e"},{"header":"Materials And Methods","content":"\u003cp\u003e\u003cstrong\u003eParticipants\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTwenty-five patients (17 males, 8 females) who had sustained moderate to severe TBI, determined prospectively using the Westmead Post Traumatic Amnesia Scale\u003csup\u003e33\u003c/sup\u003e, were recruited from the Acquired Brain Injury Ward at Epworth Hospital (Richmond, Victoria) after emerging from post-traumatic amnesia (\u003cem\u003eM =\u003c/em\u003e 2.16 months, \u003cem\u003eSD\u003c/em\u003e = 1.48 months, range = 0.69 \u0026ndash; 6.64 months; Table 1 and Supplementary Table S1). TBI patients predominantly had prefrontal and temporal pathology (Fig. 1). Exclusion criteria included age \u0026lt; 18 or \u0026gt; 75 years, prior history of TBI or other neurological conditions, significant psychiatric or substance abuse history, and MRI contraindication. Twenty-one healthy controls (13 males, 8 females) of similar age, sex, and education were also recruited (Table 1). There were no significant group differences on any of the demographic variables (\u003cem\u003eP \u003c/em\u003e\u0026gt; 0.05). Due to a technical error during data collection, a subset of the sample (6 TBIs and 8 healthy controls) was excluded from the MRI analyses due to poor coverage of the temporal lobes. Additionally, 1 healthy control participant was excluded due to excessive movement in the scanner. Despite this, it should be noted that our fMRI sample size is still comparable to previous neuroimaging studies\u003csup\u003e20-22\u003c/sup\u003e of episodic memory in the TBI population. Written informed consent was provided by all participants. This study was approved by Monash Health Human Research Ethics Committee and conducted in accordance to the Declaration of Helsinki.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEpisodic memory paradigm\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe episodic memory paradigm is a task adapted from Mundy et al. to probe episodic memory encoding and recognition (Fig. 2)\u003csup\u003e31,34\u003c/sup\u003e. Participants were presented images of faces, scenes, and animals while in the scanner. They were instructed to respond, using trigger buttons, to each stimulus based on set criteria to ensure attention was maintained throughout the task (e.g. decide whether an animal is shorter or taller than a human man; whether the face is male or female; whether a scene looks hot or cold). consisted of six blocks, each block comprising 20 images from the three stimulus categories. This task comprised 60 unique stimuli, each presented twice throughout the session. Each stimulus was presented for 3 seconds, followed by a 3 second inter-stimulus interval. Five rest blocks were presented after the 1\u003csup\u003est \u003c/sup\u003e\u0026ndash; 5\u003csup\u003eth\u003c/sup\u003e experimental blocks.\u003c/p\u003e\n\u003cp\u003eMemory of the stimuli presented in the fMRI task was probed in an out of scanner task. Recognition of the episodic stimuli was assessed by instructing participants to classify stimuli as \u0026ldquo;old\u0026rdquo; (i.e. images seen during the fMRI task) or \u0026ldquo;new\u0026rdquo; (i.e. images that were not presented during the fMRI task). Participants rated a total of 60 old and 60 new stimuli, each presented twice (240 stimuli in total). To assess retrieval \u0026lsquo;confidence\u0026rsquo;, images were rated on a scale from \u0026ldquo;definitely old\u0026rdquo; to \u0026ldquo;definitely new\u0026rdquo;. Responses were considered correct/incorrect regardless of confidence level (secondary analysis indicated no significant differences in confidence rating between groups \u0026ndash; see \u0026ldquo;Confidence rating analysis\u0026rdquo; in Supplementary for further detail). Performance for this task was assessed using accuracy and reaction times.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMRI acquisition \u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eStructural and functional MR images were acquired in two clinical scanners, using 3.0 Tesla Siemens Magnetom Skyra scanners and a 20-channel head coils. Functional images were acquired using single-shot gradient-echo planar imaging (EPI) with the following parameters: repetition time (TR) = 2.75 s; echo time (TE) = 30 ms; flip angle = 90\u0026deg;; 220 \u0026times; 220 matrix; voxel size = 3.4 \u0026times; 3.4 \u0026times; 3.0 mm. A high-resolution 3D T1-weighted image covering the entire brain was also acquired for anatomical reference (TR = 2.3 s; TE = 2.32 ms; flip angle = 8\u0026deg;; 236 \u0026times; 350 matrix; voxel size = 0.9 \u0026times; 0.9 \u0026times; 0.9 mm). Due to reduced brain coverage (due to using clinical scanners), and a-priori hypotheses, we focused on the temporal lobes.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStatistical analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eBehavioural and demographic data\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBehavioural and demographic data were analysed using R version 3.6.0 (R Core Team, 2019). Two-tailed independent samples t-tests were used to examine group differences on the demographic variables including age, sex, and years of education. Behavioural data were screened for normality, transformed (if necessary), and assessed for violation of statistical assumptions prior to analysis. Outcome measures were analysed using linear mixed models to account for clustering or non-independence of measures within participants. Memory performance was assessed using dprime and reaction time. Dprime measured task accuracy, accounting for the signal (hits) and noise (false alarms). Reaction times were generated by obtaining the average reaction time per stimulus category. Task accuracy was assessed by modelling stimulus category, group, and their interaction (stimulus category x group) as fixed effects, and participant as a random effect. For reaction time, the data were inversely transformed, and a model was fitted with stimulus category, group, and an interaction (stimulus category x group) as fixed effects, and participant as a random effect. Age and education were also added as covariates in the models, given their influence on memory performance \u003csup\u003e35,36\u003c/sup\u003e and reaction time \u003csup\u003e37\u003c/sup\u003e. Where appropriate, post-hoc analyses were conducted using two-tailed t-tests multiple comparison correction.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eImaging data\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eMRI preprocessing \u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePrior to preprocessing, lesions were manually segmented using MRIcron (\u003ca href=\"http://www.mricro.com/mricron\"\u003ehttp://www.mricro.com/mricron\u003c/a\u003e). Preprocessing was performed using fMRIPrep 20.0.0 (Esteban et al., 2018) and involved the application of the following step: undistortion of EPI data, realignment, normalisation, and estimation of confounds. Further information about the MRI preprocessing can be accessed in the Supplementary (see \u0026ldquo;Detailed MRI preprocessing\u0026rdquo;).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003efMRI analysis\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003efMRI data were analysed using FSL\u0026rsquo;s FEAT version 6.0.2 (FMRIB's Software Library, www.fmrib.ox.ac.uk/fsl). In the first level analysis, contrasts between each stimulus category (i.e. animals, faces, scenes) and the rest blocks were generated for each participant. The onset times for each contrast corresponded to the first stimulus presentation of each category and were 27 seconds in duration. To reduce motion-related artifacts, additional regressors using a modified method of the anatomical CompCor that explained 50% of the variance were also included in the first level model \u003csup\u003e38\u003c/sup\u003e. Differences in brain activation of categories of episodic stimuli were assessed in a 2 (group: TBI vs. healthy controls) x 3 (stimulus category: faces, scenes, and animals) factorial design using FLAME 1+2 mixed effects with automatic outlier de-weighting. An additional explanatory variable was also added at the group level to control for the acquisition of images from two scanners. Given our hypothesis regarding structures in the temporal lobes and their role in learning and processing of category-specific stimuli, an a-priori region of interest (ROI) mask of the temporal lobe was generated using the MNI Structural Atlas (Supplementary Fig. S1) and used in the group level analysis. Imaging findings are reported using a cluster level threshold of \u003cem\u003eZ \u003c/em\u003e\u0026gt; 3.1 and a family wise error cluster correction threshold of \u003cem\u003eP \u003c/em\u003e\u0026lt; 0.05. FEAT contrast of parameter estimates (COPE) were extracted from significant clusters at the group level. To investigate differences in BOLD response, two-tailed independent samples t-tests were conducted using COPE values. Finally, association between BOLD response and behavioural performance on the episodic recognition task were assessed using Pearson correlations.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eEpisodic memory is selectively impaired for face stimuli\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFirst, we investigated whether recognition of episodic memories was impaired following TBI. To do this, we performed a linear mixed model assessing memory recall accuracy (old vs. new, as measured using dprime) on the episodic retrieval task. Overall, the TBI group demonstrated significantly poorer recognition accuracy than healthy controls (95% CI, 0.04 \u0026ndash; 0.72; \u003cem\u003eP =\u003c/em\u003e 0.028; Fig. 3). Post-hoc analyses indicated that the group difference was driven by a significant difference in accuracy for faces (95% CI, 1.11 \u0026ndash; 1.50; \u003cem\u003eP =\u003c/em\u003e 0.037). There was a trend for lower accuracy for scenes, however, this did not reach statistical significance (95% CI, 0.94 \u0026ndash; 1. 19; \u003cem\u003eP =\u003c/em\u003e 0.169). There was no significant difference between the groups in accuracy for animals (95% CI, 1. 12 \u0026ndash; 1. 26; \u003cem\u003eP =\u003c/em\u003e 0.440).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eReaction times were quickest for face stimuli\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNext, we investigated whether reaction time was significantly different between the groups and whether it varied according to the type of stimulus. Overall, reaction time was greater for individuals with TBI compared to healthy controls, irrespective of stimulus type (95% CI, 0.08 \u0026ndash; 0.25; \u003cem\u003eP \u003c/em\u003e\u0026lt; 0.001; Fig. 3). Post-hoc analyses revealed that the TBI group was significantly slower than healthy controls in responding to faces (95% CI, 0.48 \u0026ndash; 0.64; \u003cem\u003eP \u003c/em\u003e\u0026lt; 0.001), scenes (95% CI, 0.43 \u0026ndash; 0.58; \u003cem\u003eP =\u003c/em\u003e 0.001), and animals (95% CI, 0.42 \u0026ndash; 0.58; \u003cem\u003eP =\u003c/em\u003e 0.001). Across both groups, reaction time also varied depending on the stimulus category: the TBI group was quicker to respond to faces than scenes (95% CI, 0.43 \u0026ndash; 0.48; \u003cem\u003eP \u003c/em\u003e= 0.014) and animals (95% CI, 0.42 \u0026ndash; 0.48; \u003cem\u003eP \u003c/em\u003e= 0.005); similarly, healthy controls were quicker to respond to faces than scenes (95% CI, 0.58 \u0026ndash; 0.64; \u003cem\u003eP =\u003c/em\u003e 0.003) and animals (95% CI, 0.58 \u0026ndash; 0.64; \u003cem\u003eP =\u003c/em\u003e 0.001). Given this pattern of results, we further investigated whether the poorer performance for faces in the TBI group was driven by a speed-accuracy trade-off. To do this, we included reaction time as a covariate in the linear mixed model. Results indicated that participants with TBI still performed significantly poorer than healthy controls after controlling for reaction time (95% CI, 1. 10 \u0026ndash; 1. 54; \u003cem\u003eP =\u003c/em\u003e 0.027), suggesting that this pattern of performance was not due solely to a speed-accuracy trade-off.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003efMRI Task activates the stereotypical regions underpinning encoding of episodic stimuli\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo demonstrate that our task elicited activations in stereotypical areas involved with the canonical network that support encoding of episodic stimuli, we first included all participants in an analysis looking at the average activation for each stimulus category (i.e. faces, scenes, and animals; Fig. 4). During encoding of face stimuli, significant clusters were noted in face-selective areas including the right inferior occipital gyrus and left/right fusiform gyrus\u003csup\u003e25-27\u003c/sup\u003e, as well as the right hippocampus. During encoding of scene stimuli, significant clusters were noted in scene-selective area of right parahippocampal gyrus\u003csup\u003e28\u003c/sup\u003e, as well as the left fusiform gyrus. Finally, during encoding of animal stimuli, significant clusters were noted in animal-selective area of the right fusiform gyrus\u003csup\u003e23,29\u003c/sup\u003e, as well as the left inferior occipital gyrus.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTBI patients show reduced right transverse temporal gyrus activation during face processing\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eConsistent with the behavioural results, group differences on imaging were apparent during encoding of faces. TBI participants showed reduced activation in the right transverse temporal gyrus extending to the planum temporale compared to healthy controls (Fig. 5).\u003c/p\u003e\n\u003cp\u003eFEAT analysis COPE values were extracted for the significant cluster. We first investigated whether there was a significant difference in BOLD response between groups using an independent samples t-test. As expected, TBI patients displayed lower COPE values (\u003cem\u003eM\u003c/em\u003e = -9.90, \u003cem\u003eSD \u003c/em\u003e= 24.79) compared to healthy controls (\u003cem\u003eM\u003c/em\u003e = 19.07, \u003cem\u003eSD \u003c/em\u003e= 33.04), \u003cem\u003et\u003c/em\u003e(18) = -2.61, P = 0.017. We further examined whether there was an association with behavioural performance on the episodic recognition task using Pearson correlations. Overall, there was a moderate positive relationship between COPE values and the dprime scores for face stimuli, \u003cem\u003er\u003c/em\u003e(28) = 0.497, \u003cem\u003eP =\u003c/em\u003e 0.005. Follow-up correlations indicated that a significant correlation was only apparent for the TBI group, \u003cem\u003er\u003c/em\u003e(16) = 0.480, \u003cem\u003eP =\u003c/em\u003e 0.044, and not healthy controls, \u003cem\u003er\u003c/em\u003e(10) = 0.375, \u003cem\u003eP =\u003c/em\u003e 0.229. To examine specificity of this brain-behaviour relationship, we conducted a control analysis, correlating \u003cem\u003ewhole\u003c/em\u003e temporal lobe activity with dprime scores for face stimuli. Whole temporal lobe activation was not correlated with face recognition performance (\u003cem\u003eP \u003c/em\u003e\u0026gt; 0.05; see \u0026ldquo;Additional control analyses\u0026rdquo; in Supplementary).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eTBI patients display reduced right fusiform gyrus activity during scene processing\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eDespite a non-significant difference between groups for scene recognition, we found reduced activation in the right posterior fusiform gyrus for the TBI group in comparison to healthy controls during scene encoding (Fig. 6).\u003c/p\u003e\n\u003cp\u003eTBI patients showed lower COPE values (\u003cem\u003eM\u003c/em\u003e = 29.26, \u003cem\u003eSD \u003c/em\u003e= 25.11) compared to controls (\u003cem\u003eM\u003c/em\u003e = 76.93, \u003cem\u003eSD \u003c/em\u003e= 47.76), \u003cem\u003et\u003c/em\u003e(15) = 3.18, \u003cem\u003eP\u003c/em\u003e = 0.006. To further examine whether functional changes in this cluster were associated with behavioural performance, we again conducted a series of Pearson correlations. Upon removal of an outlier (see \u0026ldquo;Additional scene cluster results\u0026rdquo; in Supplementary for further detail), there was a moderate postive relationship between the COPE values extracted from this cluster and the dprime scores for scene stimuli, \u003cem\u003er\u003c/em\u003e(27) = 0.482, \u003cem\u003eP =\u003c/em\u003e 0.008. Further examation revealed that a significant correlation was only apparent for the TBI group, \u003cem\u003er\u003c/em\u003e(15) = 0.574, \u003cem\u003eP =\u003c/em\u003e 0.016, and not healthy controls, \u003cem\u003er\u003c/em\u003e(10) = 0.317, \u003cem\u003eP =\u003c/em\u003e 0.316. Similar to the control analysis above, we conducted a control analysis, correlating \u003cem\u003ewhole\u003c/em\u003e temporal lobe activity with dprime scores for scene stimuli. Interestingly, general temporal lobe activation was correlated with scene recognition performance (\u003cem\u003eP\u003c/em\u003e = 0.006). However, correlations were no longer significant once the groups were examined separately (\u003cem\u003eP \u003c/em\u003e\u0026gt; 0.05; see \u0026ldquo;Additional control analyses\u0026rdquo; in Supplementary).\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe present study focused on determining the role of temporal lobe activity in episodic memory behaviour following TBI. We showed for the first time, using converging evidence from behavioural and fMRI data, that episodic memory impairment following TBI appeared to be category-specific and related specific sub-regions within the temporal lobes. Deficits were most apparent for faces; TBI patients displayed reduced transverse temporal gyrus activation during face encoding and subsequent impairment on face recognition. TBI patients also displayed reduced fusiform gyrus activation for scenes; this is despite no statistically significant difference between groups during scene recognition. Interestingly, brain activation during face and scene encoding correlated with subsequent recognition for TBI patients but not in healthy control participants. Overall, these findings suggest that TBI: a) preferentially impairs episodic memory in specific domains, and b) aberrant neural processing may not be reflected in statistically significant differences behavioural assessments, and thus neural activation conveys complimentary information undetected through examination of overt behaviour.\u003c/p\u003e\n\u003cp\u003eBroadly speaking, our findings are similar to those of previous studies of amnestic patients who show stimulus-sensitive impairments for complex stimuli such as faces and scenes \u003csup\u003e30,31\u003c/sup\u003e. Our findings are also in concordance with a previous study demonstrating impaired face recognition in the TBI population\u003csup\u003e32\u003c/sup\u003e. Valentine et al. subjected participants to a range of facial recognition and learning tasks and found that while performance varied, deficits were more apparent for tasks with greater demands\u003csup\u003e32\u003c/sup\u003e. More specifically, the most sensitive tasks were those which contained a larger number of faces to be encoded or had fewer presentations of the stimuli. Our task was comparably difficult in that participants were presented with a similar number of face stimuli which were only shown twice during the encoding phase; thus, it was not surprising we obtained a similar finding.\u003c/p\u003e\n\u003cp\u003eAs expected, we found that the TBI group was generally slower than healthy controls in their reaction times. Both groups, however, responded more quickly to faces than to animals and scenes. This result somewhat aligns with a study conducted by Keightley et al. who found that participants reacted quicker to faces than scenes, despite accuracy being better for scene stimuli\u003csup\u003e39\u003c/sup\u003e. We further explored whether speed-accuracy trade off could account for our findings, given participants displayed the poorest accuracy for face stimuli. We found that including reaction time as a covariate when determining between-group differences in face accuracy did not change the result. Instead, the rapid response to faces suggests that individuals with TBI may have performed superficial encoding of face stimuli, thus negatively impacting decision-making during recognition.\u003c/p\u003e\n\u003cp\u003eA unique aspect of our study was the inclusion of animal and scene stimuli, in addition to faces, which allowed us to investigate the effects of stimulus complexity on episodic memory. Interestingly, we found the TBI group displayed altered neural activity without showing statistically significant impairment of behavioural performance for scene stimuli, whereas changes on both functional and behavioural measures were apparent for faces. One potential reason for these results is that faces are more visually complex than scenes. Indeed, facial processing is a complex phenomenon requiring multifaceted processes across widespread cortical areas\u003csup\u003e40\u003c/sup\u003e. Unlike most other visual stimuli they are processed in a holistic and configural manner\u003csup\u003e41,42\u003c/sup\u003e; thus, discrimination requires attention to detail and subtle perception of variable facial features\u003csup\u003e40\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eOur fMRI results provide further insights into the mechanisms that may underlie behavioural deficits for faces. We found that the TBI group showed a reduced response in the right transverse temporal gyrus extending to the planum temporale during encoding of faces. However, these areas do not form part of the core or extended face network\u003csup\u003e26\u003c/sup\u003e. The transverse temporal gyrus is predominantly implicated in auditory processing\u003csup\u003e43,44\u003c/sup\u003e, although studies have also demonstrated its role in spontaneous inner speech\u003csup\u003e45\u003c/sup\u003e. The planum temporale has been shown to be involved in language functions\u003csup\u003e46\u003c/sup\u003e. Therefore, it is possible that difference in BOLD activity in this temporal lobe region may reflect differences in verbalisation during face encoding. Interestingly, activation of this region correlated with behavioural performance only for individuals with TBI. It may be that individuals with TBI may therefore have made less attempts to verbally describe the stimuli when processing faces. This further supports the hypothesis that TBI participants encoded these stimuli in a rapid and superficial manner. In contrast, healthy controls did not display such an association, suggesting verbalisation during face encoding was not key in determining later recognition success.\u003c/p\u003e\n\u003cp\u003eOur other key finding was that the TBI group demonstrated reduced brain activity during encoding of scenes in the right posterior fusiform gyrus. The right posterior fusiform gyrus responds non-selectively to faces and scenes and generally may be involved with processing complex visual stimuli\u003csup\u003e47\u003c/sup\u003e. This result suggests impaired recruitment of this temporal lobe sub-region during processing of scenes. Indeed, we found that individuals with TBI who had higher activation within this region performed better during scene recognition. Despite differences in brain activation, the groups did not differ with respect to behaviour. This finding may be due to scene stimuli containing a greater number of contextual cues that could further aid encoding and recollection. For example, individuals may have used cues such as the location (e.g. kitchen) or remembered certain salient scene features (e.g. item/s contained in the scene).\u003c/p\u003e\n\u003cp\u003eTo examine the specificity of the brain-behaviour relationships, we conducted a series of control analyses to determine whether the association between brain activation and behavioural performance was specific to the significant clusters or, more generally, to activity in the temporal lobes. In support of our main findings, general activation in the temporal lobes was not associated with face recognition accuracy, thus highlighting the specificity of right transverse temporal gyrus in predicting face recognition performance. However, we found that general activation in the temporal lobes was associated scene recognition accuracy. This relationship disappeared when the TBI group and healthy controls were considered separately. This indicates that the right fusiform gyrus has less specificity in predicting scene recognition performance. In line with our behaviour findings, these results suggest greater, more robust, specificity for face stimuli, compared to scenes.\u003c/p\u003e\n\u003cp\u003eAlthough our findings show a clear brain-behaviour relationship, these associations oppose our initial hypotheses, which predicted that individuals with TBI would display greater temporal lobe activity in support of previous studies\u003csup\u003e20-22\u003c/sup\u003e. One general model that could explain this finding is that of cortical reorganisation following injury\u003csup\u003e48-50\u003c/sup\u003e. That is, the pattern of cortical activation reflecting neural compensation or recovery following TBI is likely to depend on the length of time since an individual\u0026rsquo;s injury\u003csup\u003e51\u003c/sup\u003e. A key methodological difference is that our TBI participants were recruited at an average of 2 months\u0026rsquo; post-injury whereas those in past studies were recruited over 1 year post-injury{Arenth, 2012 #40}\u003csup\u003e20-22\u003c/sup\u003e{Arenth, 2012 #40}{Russell, 2011 #14}{Arenth, 2012 #40}{Arenth, 2014 #520}. In a key study, Sanchez-Carrion et al. characterised the longitudinal changes in brain activity following TBI\u003csup\u003e52\u003c/sup\u003e. They did so by assessing brain activity during a working memory task at 6 months and 1 year following the injury. Their findings show an initial reduction in brain activity at the 6 month time-points which gradually resolved by 1 year following the injury. Thus, discrepancies in our findings from past studies may reflect differences in recovery phases.\u003c/p\u003e\n\u003cp\u003eOur study has several important implications. From a clinical perspective, it is generally acknowledged that individuals have generalised episodic memory deficits after injury. Our findings provide evidence to the contrary and show that impairment is more apparent with complex visual stimuli such as faces and scenes. An obvious clinical translation is the need to provide strategies that promote deeper processing to better aid memory for these and other complex stimuli. In addition, our study provides further support for the utility of fMRI as a complimentary source of information by demonstrating evidence of aberrant neural processing that was less evident in behavioural performance. This is important considering that most assessments of memory are based on behavioural performance.\u003c/p\u003e\n\u003cp\u003eThere were some limitations in our study, however. Our episodic memory paradigm only allowed us to investigate functional activity during stimulus encoding. Therefore, we could not comment on the extent in which temporal structures are implicated in recognition of episodic stimuli following TBI. This may be an avenue for exploration in future research. Although not a limitation per se, we took an a-priori approach specifically focusing on the temporal lobes. While this allowed us to answer specific questions about the temporal lobes\u0026rsquo; role in processing of episodic stimuli, it also limited our investigation of non-temporal contributions to episodic memory. For example, there is value in investigating the interaction between frontal and temporal regions to support encoding of more complex stimuli, especially considering the frontal lobes\u0026rsquo; role in strategy, allocation of resources, and planning\u003csup\u003e53,54\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eIn conclusion, we found evidence demonstrating that individuals with TBI show impairment of episodic memory for complex stimuli and that this was associated with functional changes. In comparison to healthy controls, we found that the TBI group displayed reduced activation in the right transverse temporal gyrus and fusiform gyrus during face and scene processing, respectively. We found that brain activation in these temporal lobe sub-regions were associated with behavioural performance for the TBI group and not healthy controls. These findings may be explained in terms of differences in verbalisation during encoding and cortical reorganisation. Overall, we provide preliminary evidence demonstrating that following TBI: a) episodic memory impairment is domain specific and more broadly dependent on the complexity of the stimuli, and b) aberrant neuronal activity may exist despite lack of evidence of significant impairment in behavioural performance, and therefore neural activation may be a more robust early indicator than behaviour.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eTBI = traumatic brain injury; functional magnetic resonance imaging = fMRI; GSC = Glasgow Coma Scale; Post Traumatic Amnesia = PTA; WPTAS = Westmead Post Traumatic Amnesia Scale; DAI = diffuse axonal injury; EDH = extradural haematoma; ICH = intracerebral haemorrhage; SAH = subarachnoid haemorrhage; SDH = subdural haemorrhage; NAD = no abnormality detected; TR = repetition time; TE = echo time; contrast of parameter estimates = COPE\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGS was funded by a National Health and Research Council Early Career Fellowship (APP1104692) and the Brain Foundation. This work was supported by the Multi-modal Australian ScienceS Imaging and Visualisation Environment (MASSIVE) HPC facility (\u003ca href=\"http://www.massive.org.au\"\u003ewww.massive.org.au\u003c/a\u003e). In addition, the authors would like to thank the staff at the Acquired Brain Injury Ward at Epworth Hospital (Richmond) and Bridge Road Imaging. The authors would also like to thank the participants who took part in the study.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors contribution\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAT collected/analysed the data and wrote the manuscript; MM conceptualised the study, designed the study, and reviewed the manuscript; JP conceptualised the study and reviewed the manuscript; GS conceptualised the study, designed the study, collected/analysed the data, and reviewed the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics declaration\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompleting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll data supporting the findings of this study can be requested from the corresponding author.\u003c/p\u003e"},{"header":"References","content":"\u003cp\u003e1 Bigler, E. D. The lesion (s) in traumatic brain injury: Implications for clinical neuropsychology. \u003cem\u003eArchives of clinical neuropsychology\u003c/em\u003e\u003cstrong\u003e16\u003c/strong\u003e, 95-131 (2001).\u003c/p\u003e\n\u003cp\u003e2 Draper, K. \u0026amp; Ponsford, J. Cognitive functioning ten years following traumatic brain injury and rehabilitation. \u003cem\u003eNeuropsychology\u003c/em\u003e\u003cstrong\u003e22\u003c/strong\u003e, 618-625, doi:10.1037/0894-4105.22.5.618 (2008).\u003c/p\u003e\n\u003cp\u003e3 Azouvi, P., Arnould, A., Dromer, E. \u0026amp; Vallat-Azouvi, C. Neuropsychology of traumatic brain injury: An expert overview. \u003cem\u003eRev Neurol (Paris)\u003c/em\u003e\u003cstrong\u003e173\u003c/strong\u003e, 461-472, doi:10.1016/j.neurol.2017.07.006 (2017).\u003c/p\u003e\n\u003cp\u003e4 Tittle, A. \u0026amp; Burgess, G. H. Relative contribution of attention and memory toward disorientation or post-traumatic amnesia in an acute brain injury sample. \u003cem\u003eBrain Injury\u003c/em\u003e\u003cstrong\u003e25\u003c/strong\u003e, 933-942, doi:10.3109/02699052.2011.597042 (2011).\u003c/p\u003e\n\u003cp\u003e5 Rabinowitz, A. R. \u0026amp; Levin, H. S. Cognitive sequelae of traumatic brain injury. \u003cem\u003ePsychiatr Clin North Am\u003c/em\u003e\u003cstrong\u003e37\u003c/strong\u003e, 1-11, doi:10.1016/j.psc.2013.11.004 (2014).\u003c/p\u003e\n\u003cp\u003e6 Vakil, E. The effect of moderate to severe traumatic brain injury (TBI) on different aspects of memory: a selective review. \u003cem\u003eJ Clin Exp Neuropsychol\u003c/em\u003e\u003cstrong\u003e27\u003c/strong\u003e, 977-1021, doi:10.1080/13803390490919245 (2005).\u003c/p\u003e\n\u003cp\u003e7 Moscovitch, M., Cabeza, R., Winocur, G. \u0026amp; Nadel, L. Episodic Memory and Beyond: The Hippocampus and Neocortex in Transformation. \u003cem\u003eAnnu Rev Psychol\u003c/em\u003e\u003cstrong\u003e67\u003c/strong\u003e, 105-134, doi:10.1146/annurev-psych-113011-143733 (2016).\u003c/p\u003e\n\u003cp\u003e8 Tulving, E. Episodic memory: from mind to brain. \u003cem\u003eAnnual review of psychology\u003c/em\u003e\u003cstrong\u003e53\u003c/strong\u003e, 1-25 (2002).\u003c/p\u003e\n\u003cp\u003e9 Nakase-Richardson, R.\u003cem\u003e et al.\u003c/em\u003e Utility of post-traumatic amnesia in predicting 1-year productivity following traumatic brain injury: comparison of the Russell and Mississippi PTA classification intervals. \u003cem\u003eJ Neurol Neurosurg Psychiatry\u003c/em\u003e\u003cstrong\u003e82\u003c/strong\u003e, 494-499, doi:10.1136/jnnp.2010.222489 (2011).\u003c/p\u003e\n\u003cp\u003e10 Wagner, A. D., Shannon, B. J., Kahn, I. \u0026amp; Buckner, R. L. Parietal lobe contributions to episodic memory retrieval. \u003cem\u003eTrends Cogn Sci\u003c/em\u003e\u003cstrong\u003e9\u003c/strong\u003e, 445-453, doi:10.1016/j.tics.2005.07.001 (2005).\u003c/p\u003e\n\u003cp\u003e11 Eichenbaum, H. Prefrontal-hippocampal interactions in episodic memory. \u003cem\u003eNat Rev Neurosci\u003c/em\u003e\u003cstrong\u003e18\u003c/strong\u003e, 547-558, doi:10.1038/nrn.2017.74 (2017).\u003c/p\u003e\n\u003cp\u003e12 Dickerson, B. C. \u0026amp; Eichenbaum, H. The episodic memory system: neurocircuitry and disorders. \u003cem\u003eNeuropsychopharmacology\u003c/em\u003e\u003cstrong\u003e35\u003c/strong\u003e, 86-104, doi:10.1038/npp.2009.126 (2010).\u003c/p\u003e\n\u003cp\u003e13 Simons, J. S. \u0026amp; Spiers, H. J. Prefrontal and medial temporal lobe interactions in long-term memory. \u003cem\u003eNat Rev Neurosci\u003c/em\u003e\u003cstrong\u003e4\u003c/strong\u003e, 637-648, doi:10.1038/nrn1178 (2003).\u003c/p\u003e\n\u003cp\u003e14 Graham, K. S., Barense, M. D. \u0026amp; Lee, A. C. Going beyond LTM in the MTL: a synthesis of neuropsychological and neuroimaging findings on the role of the medial temporal lobe in memory and perception. \u003cem\u003eNeuropsychologia\u003c/em\u003e\u003cstrong\u003e48\u003c/strong\u003e, 831-853, doi:10.1016/j.neuropsychologia.2010.01.001 (2010).\u003c/p\u003e\n\u003cp\u003e15 Cameron, K. A., Yashar, S., Wilson, C. L. \u0026amp; Fried, I. Human hippocampal neurons predict how well word pairs will be remembered. \u003cem\u003eNeuron\u003c/em\u003e\u003cstrong\u003e30\u003c/strong\u003e, 289-298 (2001).\u003c/p\u003e\n\u003cp\u003e16 Barlow, K. M. Traumatic brain injury. \u003cem\u003eHandb Clin Neurol\u003c/em\u003e\u003cstrong\u003e112\u003c/strong\u003e, 891-904, doi:10.1016/B978-0-444-52910-7.00011-8 (2013).\u003c/p\u003e\n\u003cp\u003e17 Daneshvar, D. H. \u0026amp; McKee, A. C. Traumatic Brain Injury. 219-235, doi:10.1016/b978-0-12-398270-4.00016-1 (2015).\u003c/p\u003e\n\u003cp\u003e18 Bigler, E. D.\u003cem\u003e et al.\u003c/em\u003e Traumatic brain injury and memory: The role of hippocampal atrophy. \u003cem\u003eNeuropsychology\u003c/em\u003e\u003cstrong\u003e10\u003c/strong\u003e, 333 (1996).\u003c/p\u003e\n\u003cp\u003e19 Ariza, M.\u003cem\u003e et al.\u003c/em\u003e Hippocampal head atrophy after traumatic brain injury. \u003cem\u003eNeuropsychologia\u003c/em\u003e\u003cstrong\u003e44\u003c/strong\u003e, 1956-1961, doi:10.1016/j.neuropsychologia.2005.11.007 (2006).\u003c/p\u003e\n\u003cp\u003e20 Russell, K. C., Arenth, P. M., Scanlon, J. M., Kessler, L. J. \u0026amp; Ricker, J. H. A functional magnetic resonance imaging investigation of episodic memory after traumatic brain injury. \u003cem\u003eJ Clin Exp Neuropsychol\u003c/em\u003e\u003cstrong\u003e33\u003c/strong\u003e, 538-547, doi:10.1080/13803395.2010.537253 (2011).\u003c/p\u003e\n\u003cp\u003e21 Arenth, P. M., Russell, K. C., Scanlon, J. M., Kessler, L. J. \u0026amp; Ricker, J. H. Encoding and recognition after traumatic brain injury: neuropsychological and functional magnetic resonance imaging findings. \u003cem\u003eJ Clin Exp Neuropsychol\u003c/em\u003e\u003cstrong\u003e34\u003c/strong\u003e, 333-344, doi:10.1080/13803395.2011.633896 (2012).\u003c/p\u003e\n\u003cp\u003e22 Gillis, M. M. \u0026amp; Hampstead, B. M. A two-part preliminary investigation of encoding-related activation changes after moderate to severe traumatic brain injury: hyperactivation, repetition suppression, and the role of the prefrontal cortex. \u003cem\u003eBrain Imaging Behav\u003c/em\u003e\u003cstrong\u003e9\u003c/strong\u003e, 801-820, doi:10.1007/s11682-014-9337-5 (2015).\u003c/p\u003e\n\u003cp\u003e23 Downing, P. E., Chan, A. W., Peelen, M. V., Dodds, C. M. \u0026amp; Kanwisher, N. Domain specificity in visual cortex. \u003cem\u003eCereb Cortex\u003c/em\u003e\u003cstrong\u003e16\u003c/strong\u003e, 1453-1461, doi:10.1093/cercor/bhj086 (2006).\u003c/p\u003e\n\u003cp\u003e24 Mundy, M. E.\u003cem\u003e et al.\u003c/em\u003e Material-independent and material-specific activation in functional MRI after perceptual learning. \u003cem\u003eNeuroreport\u003c/em\u003e\u003cstrong\u003e20\u003c/strong\u003e, 1397-1401 (2009).\u003c/p\u003e\n\u003cp\u003e25 Hoffman, E. A. \u0026amp; Haxby, J. V. Distinct representations of eye gaze and identity in the distributed human neural system for face perception. \u003cem\u003eNature neuroscience\u003c/em\u003e\u003cstrong\u003e3\u003c/strong\u003e, 80-84 (2000).\u003c/p\u003e\n\u003cp\u003e26 Haxby, J. V., Hoffman, E. A. \u0026amp; Gobbini, M. I. Human neural systems for face recognition and social communication. \u003cem\u003eBiological psychiatry\u003c/em\u003e\u003cstrong\u003e51\u003c/strong\u003e, 59-67 (2002).\u003c/p\u003e\n\u003cp\u003e27 Kesler, M. L.\u003cem\u003e et al.\u003c/em\u003e Neural substrates of facial emotion processing using fMRI. \u003cem\u003eCognitive Brain Research\u003c/em\u003e\u003cstrong\u003e11\u003c/strong\u003e, 213-226 (2001).\u003c/p\u003e\n\u003cp\u003e28 Epstein, R. A. \u0026amp; Ward, E. J. How reliable are visual context effects in the parahippocampal place area? \u003cem\u003eCereb Cortex\u003c/em\u003e\u003cstrong\u003e20\u003c/strong\u003e, 294-303, doi:10.1093/cercor/bhp099 (2010).\u003c/p\u003e\n\u003cp\u003e29 Rogers, T. T., Hocking, J., Mechelli, A., Patterson, K. \u0026amp; Price, C. Fusiform activation to animals is driven by the process, not the stimulus. \u003cem\u003eJournal of Cognitive Neuroscience\u003c/em\u003e\u003cstrong\u003e17\u003c/strong\u003e, 434-445 (2005).\u003c/p\u003e\n\u003cp\u003e30 Taylor, K. J., Henson, R. N. \u0026amp; Graham, K. S. Recognition memory for faces and scenes in amnesia: dissociable roles of medial temporal lobe structures. \u003cem\u003eNeuropsychologia\u003c/em\u003e\u003cstrong\u003e45\u003c/strong\u003e, 2428-2438, doi:10.1016/j.neuropsychologia.2007.04.004 (2007).\u003c/p\u003e\n\u003cp\u003e31 Mundy, M. E., Downing, P. E., Dwyer, D. M., Honey, R. C. \u0026amp; Graham, K. S. A critical role for the hippocampus and perirhinal cortex in perceptual learning of scenes and faces: complementary findings from amnesia and FMRI. \u003cem\u003eJournal of Neuroscience\u003c/em\u003e\u003cstrong\u003e33\u003c/strong\u003e, 10490-10502 (2013).\u003c/p\u003e\n\u003cp\u003e32 Valentine, T., Powell, J., Davidoff, J., Letson, S. \u0026amp; Greenwood, R. Prevalence and correlates of face recognition impairments after acquired brain injury. \u003cem\u003eNeuropsychol Rehabil\u003c/em\u003e\u003cstrong\u003e16\u003c/strong\u003e, 272-297, doi:10.1080/09602010500176443 (2006).\u003c/p\u003e\n\u003cp\u003e33 Shores, E. A., Marosszeky, J., Sandanam, J. \u0026amp; Batchelor, J. Preliminary validation of a clinical scale for measuring the duration of post-traumatic amnesia. \u003cem\u003eMed J Aust\u003c/em\u003e\u003cstrong\u003e144\u003c/strong\u003e, 569-572 (1986).\u003c/p\u003e\n\u003cp\u003e34 Mundy, M. E., Downing, P. E. \u0026amp; Graham, K. S. Extrastriate cortex and medial temporal lobe regions respond differentially to visual feature overlap within preferred stimulus category. \u003cem\u003eNeuropsychologia\u003c/em\u003e\u003cstrong\u003e50\u003c/strong\u003e, 3053-3061 (2012).\u003c/p\u003e\n\u003cp\u003e35 Hoyer, W. J. \u0026amp; Verhaeghen, P. in \u003cem\u003eHandbook of the psychology of aging\u003c/em\u003e 209-232 (Elsevier, 2006).\u003c/p\u003e\n\u003cp\u003e36 Lachman, M. E., Agrigoroaei, S., Murphy, C. \u0026amp; Tun, P. A. Frequent cognitive activity compensates for education differences in episodic memory. \u003cem\u003eThe American Journal of Geriatric Psychiatry\u003c/em\u003e\u003cstrong\u003e18\u003c/strong\u003e, 4-10 (2010).\u003c/p\u003e\n\u003cp\u003e37 Der, G. \u0026amp; Deary, I. J. The relationship between intelligence and reaction time varies with age: Results from three representative narrow-age age cohorts at 30, 50 and 69 years. \u003cem\u003eIntelligence\u003c/em\u003e\u003cstrong\u003e64\u003c/strong\u003e, 89-97 (2017).\u003c/p\u003e\n\u003cp\u003e38 Muschelli, J.\u003cem\u003e et al.\u003c/em\u003e Reduction of motion-related artifacts in resting state fMRI using aCompCor. \u003cem\u003eNeuroimage\u003c/em\u003e\u003cstrong\u003e96\u003c/strong\u003e, 22-35, doi:10.1016/j.neuroimage.2014.03.028 (2014).\u003c/p\u003e\n\u003cp\u003e39 Keightley, M. L., Chiew, K. S., Anderson, J. A. \u0026amp; Grady, C. L. Neural correlates of recognition memory for emotional faces and scenes. \u003cem\u003eSoc Cogn Affect Neurosci\u003c/em\u003e\u003cstrong\u003e6\u003c/strong\u003e, 24-37, doi:10.1093/scan/nsq003 (2011).\u003c/p\u003e\n\u003cp\u003e40 Haxby, J. V., Hoffman, E. A. \u0026amp; Gobbini, M. I. The distributed human neural system for face perception. \u003cem\u003eTrends Cogn. Sci.\u003c/em\u003e\u003cstrong\u003e4\u003c/strong\u003e, 223-233 (2000).\u003c/p\u003e\n\u003cp\u003e41 Park, J., Newman, L. I. \u0026amp; Polk, T. A. Face processing: the interplay of nature and nurture. \u003cem\u003eThe Neuroscientist\u003c/em\u003e\u003cstrong\u003e15\u003c/strong\u003e, 445-449 (2009).\u003c/p\u003e\n\u003cp\u003e42 Maurer, D., Le Grand, R. \u0026amp; Mondloch, C. J. The many faces of configural processing. \u003cem\u003eTrends Cogn. Sci.\u003c/em\u003e\u003cstrong\u003e6\u003c/strong\u003e, 255-260 (2002).\u003c/p\u003e\n\u003cp\u003e43 Kaas, J. H., Hackett, T. A. \u0026amp; Tramo, M. J. Auditory processing in primate cerebral cortex. \u003cem\u003eCurr. Opin. Neurobiol.\u003c/em\u003e\u003cstrong\u003e9\u003c/strong\u003e, 164-170 (1999).\u003c/p\u003e\n\u003cp\u003e44 Warrier, C.\u003cem\u003e et al.\u003c/em\u003e Relating structure to function: Heschl's gyrus and acoustic processing. \u003cem\u003eJ. Neurosci.\u003c/em\u003e\u003cstrong\u003e29\u003c/strong\u003e, 61-69 (2009).\u003c/p\u003e\n\u003cp\u003e45 Hurlburt, R. T., Alderson-Day, B., K\u0026uuml;hn, S. \u0026amp; Fernyhough, C. Exploring the ecological validity of thinking on demand: neural correlates of elicited vs. spontaneously occurring inner speech. \u003cem\u003ePloS one\u003c/em\u003e\u003cstrong\u003e11\u003c/strong\u003e, e0147932 (2016).\u003c/p\u003e\n\u003cp\u003e46 Shapleske, J., Rossell, S. L., Woodruff, P. \u0026amp; David, A. The planum temporale: a systematic, quantitative review of its structural, functional and clinical significance. \u003cem\u003eBrain Research Reviews\u003c/em\u003e\u003cstrong\u003e29\u003c/strong\u003e, 26-49 (1999).\u003c/p\u003e\n\u003cp\u003e47 Nakamura, K.\u003cem\u003e et al.\u003c/em\u003e Functional delineation of the human occipito-temporal areas related to face and scene processing: a PET study. \u003cem\u003eBrain\u003c/em\u003e\u003cstrong\u003e123\u003c/strong\u003e, 1903-1912 (2000).\u003c/p\u003e\n\u003cp\u003e48 Levine, B.\u003cem\u003e et al.\u003c/em\u003e Functional reorganisation of memory after traumatic brain injury: a study with H2150 positron emission tomography. \u003cem\u003eJournal of Neurology, Neurosurgery \u0026amp; Psychiatry\u003c/em\u003e\u003cstrong\u003e73\u003c/strong\u003e, 173-181 (2002).\u003c/p\u003e\n\u003cp\u003e49 Christodoulou, C.\u003cem\u003e et al.\u003c/em\u003e Functional magnetic resonance imaging of working memory impairment after traumatic brain injury. \u003cem\u003eJournal of Neurology, Neurosurgery \u0026amp; Psychiatry\u003c/em\u003e\u003cstrong\u003e71\u003c/strong\u003e, 161-168 (2001).\u003c/p\u003e\n\u003cp\u003e50 Hillary, F. G. Neuroimaging of working memory dysfunction and the dilemma with brain reorganization hypotheses. \u003cem\u003eJ Int Neuropsychol Soc\u003c/em\u003e\u003cstrong\u003e14\u003c/strong\u003e, 526-534, doi:10.1017/S1355617708080788 (2008).\u003c/p\u003e\n\u003cp\u003e51 Munoz-Cespedes, J. M., Rios-Lago, M., Paul, N. \u0026amp; Maestu, F. Functional neuroimaging studies of cognitive recovery after acquired brain damage in adults. \u003cem\u003eNeuropsychol Rev\u003c/em\u003e\u003cstrong\u003e15\u003c/strong\u003e, 169-183, doi:10.1007/s11065-005-9178-5 (2005).\u003c/p\u003e\n\u003cp\u003e52 Sanchez-Carrion, R.\u003cem\u003e et al.\u003c/em\u003e A longitudinal fMRI study of working memory in severe TBI patients with diffuse axonal injury. \u003cem\u003eNeuroimage\u003c/em\u003e\u003cstrong\u003e43\u003c/strong\u003e, 421-429, doi:10.1016/j.neuroimage.2008.08.003 (2008).\u003c/p\u003e\n\u003cp\u003e53 Vakil, E., Greenstein, Y., Weiss, I. \u0026amp; Shtein, S. The Effects of Moderate-to-Severe Traumatic Brain Injury on Episodic Memory: a Meta-Analysis. \u003cem\u003eNeuropsychol Rev\u003c/em\u003e\u003cstrong\u003e29\u003c/strong\u003e, 270-287, doi:10.1007/s11065-019-09413-8 (2019).\u003c/p\u003e\n\u003cp\u003e54 Stuss, D. T. \u0026amp; Alexander, M. P. Does damage to the frontal lobes produce impairment in memory? \u003cem\u003eCurrent Directions in Psychological Science\u003c/em\u003e\u003cstrong\u003e14\u003c/strong\u003e, 84-88 (2005).\u003c/p\u003e"},{"header":"Tables","content":"\u003cp\u003eTable 1. Demographic information and clinical characteristics of the TBI and healthy groups.\u003c/p\u003e\n\u003ctable border=\"1\" width=\"80%\"\u003e\n\u003ctbody\u003e\n\u003ctr\u003e\n\u003ctd width=\"45%\"\u003e\n\u003cp\u003e\u003cstrong\u003eDemographic variables\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd width=\"30%\"\u003e\n\u003cp\u003e\u003cstrong\u003eTraumatic brain injury \u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMean (SD)\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd width=\"24%\"\u003e\n\u003cp\u003e\u003cstrong\u003eHealthy controls, \u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMean (SD)\u003c/strong\u003e\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd width=\"45%\"\u003e\n\u003cp\u003eAge (years)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd width=\"30%\"\u003e\n\u003cp\u003e38.36 (16.82)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd width=\"24%\"\u003e\n\u003cp\u003e37.29 (17.89)\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd width=\"45%\"\u003e\n\u003cp\u003eSex (male/female)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd width=\"30%\"\u003e\n\u003cp\u003e17/8\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd width=\"24%\"\u003e\n\u003cp\u003e13/8\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd width=\"45%\"\u003e\n\u003cp\u003eEducation (years)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd width=\"30%\"\u003e\n\u003cp\u003e13.82 (3.27)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd width=\"24%\"\u003e\n\u003cp\u003e14.19 (2.57)\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd width=\"45%\"\u003e\n\u003cp\u003eTime since injury (months)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd width=\"30%\"\u003e\n\u003cp\u003e2.16 (1.51)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd width=\"24%\"\u003e\n\u003cp\u003e-\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd width=\"45%\"\u003e\n\u003cp\u003ePTA (days)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd width=\"30%\"\u003e\n\u003cp\u003e22.46 (14.63)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd width=\"24%\"\u003e\n\u003cp\u003e-\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd width=\"45%\"\u003e\n\u003cp\u003eGCS (lowest)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd width=\"30%\"\u003e\n\u003cp\u003e9 (4.37)\u003c/p\u003e\n\u003c/td\u003e\n\u003ctd width=\"24%\"\u003e\n\u003cp\u003e-\u003c/p\u003e\n\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eGSC = Glasgow Coma Scale; PTA = post-traumatic amnesia. Note: PTA duration were available for \u003cem\u003en\u003c/em\u003e = 24 patients.\u003c/p\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":"traumatic brain injury, episodic memory, MRI, temporal lobes","lastPublishedDoi":"10.21203/rs.3.rs-152808/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-152808/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe temporal lobes are critical for episodic memories and are preferentially affected following a traumatic brain injury (TBI). As such, episodic memory difficulties are common following TBI; however, the underlying neural changes that precipitate or maintain these difficulties in the early phase of recovery remains poorly understood. Here, we use functional magnetic resonance imaging (fMRI) to interrogate the relationship of temporal lobe activation in response to face, scene, and animal stimuli. Twenty-five patients with moderate to severe TBI were recruited an average of 2 months’ post-injury and compared with 21 demographically similar healthy controls. Findings indicate that memory for faces was preferentially impaired, compared to scene and animal stimuli. Decreased activity in temporal lobe structures was present for both face (right transverse temporal gyrus) and scene stimuli (right fusiform gyrus), but not for animals. Greater activation in these structures was associated with better long-term recognition. These findings provide evidence to suggest that TBI: a) preferentially affects memory for complex stimuli such as faces and scenes, and b) causes aberrant neuronal processes despite lack of evidence of significant impairment in behavioural performance. The mechanisms underpinning these findings are discussed in terms of differences in verbalisation during encoding and reduced neural efficiency.\u003c/p\u003e","manuscriptTitle":"Temporal Lobe Activation Predicts Episodic Memory Following Traumatic Brain Injury","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2021-01-28 19:16:35","doi":"10.21203/rs.3.rs-152808/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Major revision","date":"2021-04-21T06:26:01+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2021-03-19T14:27:26+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"9eeb9944-97a6-48ca-bd7f-60a996e9821a","date":"2021-03-15T23:01:12+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2021-03-14T08:36:30+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2021-03-11T05:58:28+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2021-01-28T05:36:03+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2021-01-25T19:03:09+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2021-01-21T23:43:24+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":"bd94b6e5-c804-43bb-ba9e-53efa3d60f18","owner":[],"postedDate":"January 28th, 2021","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":2130897,"name":"Psychology"},{"id":2130898,"name":"Cognitive Neuroscience"}],"tags":[],"updatedAt":"2021-08-30T04:29:05+00:00","versionOfRecord":[],"versionCreatedAt":"2021-01-28 19:16:35","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-152808","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-152808","identity":"rs-152808","version":["v1"]},"buildId":"uwybb5PU2iWlRI8EIam5Y","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}