Associative Learning Research Articles

Associative Learning in Subclinical Hypothyroidism at Different Ranges of Thyroid Stimulating Hormone: A Cambridge Neuropsychological Test Automated Battery (CANTAB) Study of Visual Paired Association Learning Task.

Associative learning deficits are constantly found in subclinical hypothyroidism (SCH). Despite achieving normal thyroid stimulating hormone (TSH) levels, a considerable number of patients undergoing levothyroxine (LT-4) treatment frequently complain about memory retrieval. The Paired Association Learning (PAL) task involves computerised testing on the CANTAB- Cambridge Neuropsychological Test Automated Battery, also considered a screening tool for Alzheimer's disease (AD). This study aimed to investigate the impact of different levels of TSH on visual associative learning in SCH and determine if these impairments were reversed with LT-4. A total of 134 participants were included in this cross-sectional study. Group 1: 35 healthy controls; patients with SCH (Group 2: 33 newly identified cases; Group 3: 32 patients on LT-4 with elevated TSH; Group 4: 34 euthyroid but on LT-4). A thyroid profile and a neuropsychological clinical assessment were done. The visual PAL task was performed on CANTAB. PAL was significantly impaired (p = <0.05) in all 3 patient groups as compared to Group 1. The PAL total errors (adjusted) scores were significantly higher in Groups 2 and 3, indicating that associative learning is definitely impaired in SCH, reaching levels previously seen in patients with AD. Our findings encourage screening for visual associative learning or memory retrieval in patients with SCH. The study present has established a more reasonable threshold of TSH 2.5mIU/L to encourage examination of associative learning and the initiation of LT-4 in SCH. Poor PAL task performance in patients with SCH may have significant implications in clinical settings for suspecting AD.

Associative Learning and Active Inference.

Associative learning is a behavioral phenomenon in which individuals develop connections between stimuli or events based on their co-occurrence. Initially studied by Pavlov in his conditioning experiments, the fundamental principles of learning have been expanded on through the discovery of a wide range of learning phenomena. Computational models have been developed based on the concept of minimizing reward prediction errors. The Rescorla-Wagner model, in particular, is a well-known model that has greatly influenced the field of reinforcement learning. However, the simplicity of these models restricts their ability to fully explain the diverse range of behavioral phenomena associated with learning. In this study, we adopt the free energy principle, which suggests that living systems strive to minimize surprise or uncertainty under their internal models of the world. We consider the learning process as the minimization of free energy and investigate its relationship with the Rescorla-Wagner model, focusing on the informational aspects of learning, different types of surprise, and prediction errors based on beliefs and values. Furthermore, we explore how well-known behavioral phenomena such as blocking, overshadowing, and latent inhibition can be modeled within the active inference framework. We accomplish this by using the informational and novelty aspects of attention, which share similar ideas proposed by seemingly contradictory models such as Mackintosh and Pearce-Hall models. Thus, we demonstrate that the free energy principle, as a theoretical framework derived from first principles, can integrate the ideas and models of associative learning proposed based on empirical experiments and serve as a framework for a better understanding of the computational processes behind associative learning in the brain.

Personality may modulate learning and memory differences in two taxa of the African striped mouse genus Rhabdomys

AbstractDifferent environments place different cognitive demands on constituent taxa. Learning and memory involve cognitive processes with associated costs, and it is expected that different levels of learning will occur in taxa from different environments. Greater memory loads linked to increased environmental complexity require greater learning and memory capacities. We investigated the variation in learning and memory in sister taxa of striped mice (genus Rhabdomys). We studied two populations each of the mesic grassland‐occurring R. d. chakae and the mostly arid‐occurring R. pumilio. We conducted two sets of experiments. (1) In a novel object recognition (NOR) test, we assessed memory by recording the duration of exploration of similar and novel objects by test mice. (2) In an associative learning task, we assessed whether mice could associate specific scents with or without a food incentive or with different quantities of the food incentive in previous training phases. We measured the latency of mice to contact scents in a two‐sample choice in the test phase. In the NOR test, R. pumilio spent less time investigating similar objects in a training trial than R. d. chakae but increased absolute exploration of the novel object when presented with a novel and a familiar object in the retention trial, suggesting a sensitization to the novel object by R. pumilio. In the associative learning experiments, R. pumilio approached the stimuli faster than R. d. chakae, whereas mice from both taxa preferred scents associated with a seed versus no seed and scents associated with 5 seeds versus 1 seed. The data provide evidence of taxon‐level differences in learning and memory, likely related to environmentally modulated personality differences between the taxa.

Mitochondrial DNA copy number associated dementia risk by somatic mutations and frailty.

Mitochondrial dysfunction is linked to physical impairment and dementia. Mitochondrial DNA copy number (mtDNAcn) from blood may predict cognitive decline and dementia risk, but the effect of somatic mutations or frailty is unknown. We estimated mtDNAcn using fastMitoCalc and microheteroplasmies using mitoCaller, from Whole Genome Sequencing (WGS) data. In 189,566 participants free of dementia at study entry (mean age = 56 ± 8), we examined the association between mtDNAcn and subsequent dementia diagnosis using Cox regression. Cognition was assessed in a subset on average 8.9years later. We examined the associations between mtDNAcn and cognitive measures using multivariable linear regression, adjusted for demographic factors, mtDNAcn-related parameters, and apolipoprotein E ε4 status. We further stratified by frailty and microheteroplasmies. Over an average follow-up of 13.2years, 3533 participants developed dementia. Each SD higher mtDNAcn (16) was associated with 4.2% lower all-cause dementia hazard (HR = 0.958, p = 0.030), 6% lower non-AD dementia hazard (HR = 0.933, p = 0.022), and not-AD dementia hazard. The associations between mtDNAcn and all-cause dementia and non-AD dementia were stronger among those who were pre-frail or frail or with higher microheteroplasmies. Higher mtDNAcn was associated with higher DSST scores (p = 0.036) and significant only among those with higher microheteroplasmies or frailty (p = 0.029 and 0.048, respectively). mtDNAcn was also associated with delta TMT and paired associate learning only in pre-frail/frail participants (p = 0.007 and 0.045, respectively). Higher WGS-based mtDNAcn in human blood is associated with lower dementia risk, specifically non-AD dementia, and specific cognitive function. The relationships appear stronger in high somatic mutations or frailty. Future studies are warranted to investigate biological underpinnings.

Non-associative potentiation of proximal excitatory inputs to layer 2/3 pyramidal cells in rat visual cortex

Long-term changes of synaptic transmission can be induced by Hebbian-type homosynaptic mechanisms which require activation of both pre- and postsynapse and mediate associative learning, as well as by heterosynaptic mechanisms which do not require activation of the presynapse and are non-associative. The rules for induction of homosynaptic plasticity depend on the distance of the synapse from the soma. Does induction of heterosynaptic plasticity also depend on synaptic location? Here, we investigated heterosynaptic changes in pharmacologically isolated glutamatergic inputs arriving at either the proximal or the distal segments of the apical dendrite of layer 2/3 pyramidal neurons in rat visual cortex. We show that bursts of action potentials evoked without presynaptic stimulation induced potentiation of proximal inputs while having little effect on distal inputs. Such gradient of plasticity could be related to the attenuation of backpropagating action potentials along the dendrites. Thus, the location of the synapse on the dendritic tree is a determinant not only for homosynaptic but also for heterosynaptic plasticity.

Contributions of associative and non-associative learning to the dynamics of defensive ethograms.

Defensive behavior changes based on threat intensity, proximity, and context of exposure, and learning about danger-predicting stimuli is critical for survival. However, most Pavlovian fear conditioning paradigms focus only on freezing behavior, obscuring the contributions of associative and non-associative mechanisms to dynamic defensive responses. To thoroughly investigate defensive ethograms, we subjected male and female adult C57BL/6J mice to a Pavlovian conditioning paradigm that paired footshock with a serial compound stimulus (SCS) consisting of distinct tone and white noise (WN) stimulus periods. To investigate how associative and non-associative mechanisms affect defensive responses, we compared this paired SCS-footshock group with four control groups that were conditioned with either pseudorandom unpaired presentations of SCS and footshock, shock only, or reversed SCS presentations with inverted tone--WN order, with paired or unpaired presentations. On day 2 of conditioning, the paired group exhibited robust freezing during the tone period with switching to explosive jumping and darting behaviors during the WN period. Comparatively, the unpaired and both reverse SCS groups expressed less tone-induced freezing and rarely showed jumping or darting during WN. Following the second day of conditioning, we observed how defensive behavior changed over two extinction sessions. During extinction, the tone-induced freezing decreased in the paired group and mice rapidly shifted from escape jumping during WN to a combination of freezing and darting. The unpaired, unpaired reverse, and shock-only groups displayed defensive tail rattling and darting during the SCS, with minimal freezing and jumping. Interestingly, the paired reverse group did not jump to WN, and tone-evoked freezing was resistant to extinction. These findings demonstrate that non-associative factors promote some defensive responsiveness, but associative factors are required for robust cue-induced freezing and high-intensity flight expression.

Does bumblebee preference of continuous over interrupted strings in string-pulling tasks indicate means-end comprehension?

Bumblebees (Bombus terrestris) have been shown to engage in string-pulling behavior to access rewards. The objective of this study was to elucidate whether bumblebees display means-end comprehension in a string-pulling task. We presented bumblebees with two options: one where a string was connected to an artificial flower containing a reward and the other presenting an interrupted string. Bumblebees displayed a consistent preference for pulling connected strings over interrupted ones after training with a stepwise pulling technique. When exposed to novel string colors, bees continued to exhibit a bias towards pulling the connected string. This suggests that bumblebees engage in featural generalization of the visual display of the string connected to the flower in this task. If the view of the string connected to the flower was restricted during the training phase, the proportion of bumblebees choosing the connected strings significantly decreased. Similarly, when the bumblebees were confronted with coiled connected strings during the testing phase, they failed to identify and reject the interrupted strings. This finding underscores the significance of visual consistency in enabling the bumblebees to perform the task successfully. Our results suggest that bumblebees' ability to distinguish between continuous strings and interrupted strings relies on a combination of image matching and associative learning, rather than means-end understanding. These insights contribute to a deeper understanding of the cognitive processes employed by bumblebees when tackling complex spatial tasks.

Does bumblebee preference of continuous over interrupted strings in string-pulling tasks indicate means-end comprehension?

Bumblebees (Bombus terrestris) have been shown to engage in string-pulling behavior to access rewards. The objective of this study was to elucidate whether bumblebees display means-end comprehension in a string-pulling task. We presented bumblebees with two options: one where a string was connected to an artificial flower containing a reward and the other presenting an interrupted string. Bumblebees displayed a consistent preference for pulling connected strings over interrupted ones after training with a stepwise pulling technique. When exposed to novel string colors, bees continued to exhibit a bias towards pulling the connected string. This suggests that bumblebees engage in featural generalization of the visual display of the string connected to the flower in this task. If the view of the string connected to the flower was restricted during the training phase, the proportion of bumblebees choosing the connected strings significantly decreased. Similarly, when the bumblebees were confronted with coiled connected strings during the testing phase, they failed to identify and reject the interrupted strings. This finding underscores the significance of visual consistency in enabling the bumblebees to perform the task successfully. Our results suggest that bumblebees’ ability to distinguish between continuous strings and interrupted strings relies on a combination of image matching and associative learning, rather than means-end understanding. These insights contribute to a deeper understanding of the cognitive processes employed by bumblebees when tackling complex spatial tasks.

Formation and retrieval of cell assemblies in a biologically realistic spiking neural network model of area CA3 in the mouse hippocampus

The hippocampal formation is critical for episodic memory, with area Cornu Ammonis 3 (CA3) a necessary substrate for auto-associative pattern completion. Recent theoretical and experimental evidence suggests that the formation and retrieval of cell assemblies enable these functions. Yet, how cell assemblies are formed and retrieved in a full-scale spiking neural network (SNN) of CA3 that incorporates the observed diversity of neurons and connections within this circuit is not well understood. Here, we demonstrate that a data-driven SNN model quantitatively reflecting the neuron type-specific population sizes, intrinsic electrophysiology, connectivity statistics, synaptic signaling, and long-term plasticity of the mouse CA3 is capable of robust auto-association and pattern completion via cell assemblies. Our results show that a broad range of assembly sizes could successfully and systematically retrieve patterns from heavily incomplete or corrupted cues after a limited number of presentations. Furthermore, performance was robust with respect to partial overlap of assemblies through shared cells, substantially enhancing memory capacity. These novel findings provide computational evidence that the specific biological properties of the CA3 circuit produce an effective neural substrate for associative learning in the mammalian brain.

The ventral tegmental area dopamine to basolateral amygdala projection supports acquisition of cocaine self-administration

Dopamine signaling in the amygdala is known to play a role in associative learning and memory, including the process of learning to associate environmental cues with the reinforcing properties of drugs like cocaine. Evidence suggests that the ventral tegmental area (VTA) dopamine (DA) projection specifically to the basolateral amygdala (BLA) participates in establishing cocaine-cue associations that can promote later craving- and relapse-like responses to the cue alone. In order to further investigate the specific role of VTA-BLA projections in cocaine-reinforced learning, we used chemogenetics to manipulate VTA DA inputs to the BLA during cocaine self-administration, cue- and cocaine-primed reinstatement, and conditioned place preference. We found inhibiting DA input to the BLA during cocaine self-administration inhibited acquisition and weakened the ability of the previously cocaine-paired cue to elicit cocaine-seeking, while acutely inhibiting the pathway on the day of cue-induced reinstatement testing had no effect. Conversely, exciting the projection during self-administration boosted the salience of the cocaine-paired cue as indicated by enhanced responding during cue-induced reinstatement. Importantly, interfering with DA input to the BLA had no impact on the ability of cocaine to elicit a place preference or induce reinstatement in response to a priming cocaine injection. Overall, we show that manipulation of projections underlying DA signaling in the BLA may be useful for developing therapeutic interventions for substance use disorders.

Looming stimuli reliably drive innate defensive responses in male rats, but not learned defensive responses

Survival relies on an organism's intrinsic ability to instinctively react to stimuli such as food, water, and threats, ensuring the fundamental ability to feed, drink, and avoid danger even in the absence of prior experience. These natural, unconditioned stimuli can also facilitate associative learning, where pairing them consistently with neutral cues will elicit responses to these cues. Threat conditioning, a well-explored form of associative learning, commonly employs painful electric shocks, mimicking injury, as unconditioned stimuli. It remains elusive whether actual injury or pain is necessary for effective learning, or whether the threat of harm is sufficient. Approaching predators create looming shadows and sounds, triggering strong innate defensive responses like escape and freezing. This study investigates whether visual looming stimuli can induce learned freezing or learned escape responses to a conditioned stimulus in male rats. Surprisingly, pairing a neutral tone with a looming stimulus only weakly evokes learned defensive responses, in contrast to the strong responses observed when the looming stimulus is replaced by a shock. This dissociation sheds light on the boundaries for learned defensive responses thereby impacting our comprehension of learning processes and defensive strategies.

Deleterious effect of social instability on broiler chicken learning abilities and behaviour

Broiler chickens are reared in very large groups and are submitted to repeated encounters with unknown conspecifics. Our aim was to assess the consequences of these encounters on broiler chickens learning abilities, social behaviour and emotional reactivity. Groups of 5 male medium-growing broiler chickens (JA 957) were reared under stable or unstable social conditions (n=16 groups in each condition). Under the unstable conditions, the 5 birds of the groups changed every 3–4 days from 1 to 52 days of age. Broiler chickens were tested in an associative learning task (conditioned place preference) between 14 and 18 days of age. A highly palatable food (mealworms) was delivered in a particular environment with coloured stripes and preference for this environment was then tested. Social behaviours were analysed between 21 and 43 days of age. Emotional reactivity of the birds when faced with novelty (reactivity to unknown food, object and human) was assessed at the end of the rearing period (46–51 days of age). Only birds from the stable condition were able to associate the palatable food to the environment in which it was delivered (p = 0.02). Furthermore, these birds had a higher social proximity than those from the unstable condition (p < 0.01) and showed less aggressive interactions such as aggressive pecking (p = 0.06), threat (p < 0.01) or facing each other with wing flapping in front of another bird (p = 0.03). The social conditions tested had no significant effect on their emotional reactivity. Our results evidenced that chickens exposed to social instability have some learning abilities impaired compared to chickens maintained in stable social conditions. In addition, social instability increases aggressive interactions between conspecifics, while social stability favours bird proximity. In conclusion, being reared in unstable social conditions likely impairs bird adaptation to their rearing environment and increases deleterious consequences of aggressive interactions with conspecifics.

B - 115 How Training Conditions Impact Performance on a Theoretically-Derived Learning Instrument

Abstract Objective Few neuropsychological measures were developed using an explicit theory that accounts for individual performance. We developed a novel associative learning measure to address this limitation and investigated whether training conditions could influence accuracy and response time. Methods Fifty young, community-dwelling adult participants completed the novel relational learning task, which explicitly trained two stimulus pairs (A- &amp;gt; B, A- &amp;gt; C) and involved a single common stimulus. Participants identify six unique relationships between stimuli: trained relations (A- &amp;gt; B, A- &amp;gt; C), mutually entailed relations (B- &amp;gt; A, C- &amp;gt; A), and combinatorial relations (B- &amp;gt; C, C- &amp;gt; B). Participants were randomly assigned to a congruent (A is opposite B, A is opposite C) or incongruent (A is opposite B, B is the same as C) training condition and completed the procedure when stimuli were observably (e.g., color-coded shapes) or arbitrarily related (e.g., letters). Results Comparisons of accuracy between training conditions using Bayesian independent samples t-tests found moderate evidence of no reliable difference between training groups when stimuli were observably related (BF01 = 3.4, δ = 0.06) and arbitrarily related (BF01 = 3.5, δ = 0.03). Comparisons of response time found strong evidence that participants took longer to identify relations when receiving incongruent training when the stimuli were visually related (BF10 = 38.8, δ = 0.9) but inconclusive evidence when the stimuli were arbitrarily related (BF10 = 1.5, δ = 0.49) (See Table 1). Conclusion These findings suggest that manipulating training conditions does not appreciably impact the ability to identify relationships between stimuli accurately. However, training appears to increase the response time needed to identify visibly related stimuli accurately. While this is a promising theoretically-derived learning paradigm, further research is required for validation in clinical practice.

Learning motivation of college students in multimedia environment with machine learning models

As the main force of higher education, ensuring the learning status and quality of college students is undoubtedly an important task in the education industry. Analyzing their learning motivation can provide a good understanding of their learning status. Especially in the new educational environment supported by multimedia technology, efficient and convenient learning channels can eliminate students' concerns about educational facilities and instead strengthen the analysis of learning motivation in other aspects. As part of our comprehensive study of learning motivation, we draw on established learning theories, such as reinforcement theory, associative learning, and self-determination theory. Applying such learning theories encourages positive reinforcement, establishes constructive relationships with learning, and nurtures competence and autonomy. This article believed that using machine learning models to predict students' grades or behaviours and analyze their learning motivation is a good approach. Moreover, this article also tested the prediction accuracy by setting different improved random forest model runs, and concluded that the more runs, the higher the accuracy. Especially when the runs reached 100, the accuracy reached 99.98 %.

Multimodal assessment of the role of intolerance of uncertainty in fear acquisition and extinction

Personality traits linked to internalizing disorders influence the way we develop fears, but also how we regain a sense of safety. In the present study, we investigated the effect of intolerance of uncertainty (IU) on defensive responses using a differential fear conditioning protocol with an extinction phase. The conditioned stimulus was associated with an aversive sound (90 dB) in 75 % of the presentations during acquisition. A final sample of 176 participants completed the experiment. We measured self-reports of associative (expectancy of the unconditioned stimulus in acquisition) and evaluative learning (arousal and valence), and both physiological (skin conductance response) and electrocortical responses (steady-state visually evoked potentials, ssVEPs; late positive potentials, LPP) to the conditioned stimuli. Our results show that IU's impact is limited, with no effect in both acquisition and extinction. These findings emphasize the necessity of large samples in research on inter-individual differences and contribute to our understanding of how IU may or may not be involved in fear and safety learning processes considering multiple aspects of fear responding.

Reinforcement learning as a robotics-inspired framework for insect navigation: from spatial representations to neural implementation.

Bees are among the master navigators of the insect world. Despite impressive advances in robot navigation research, the performance of these insects is still unrivaled by any artificial system in terms of training efficiency and generalization capabilities, particularly considering the limited computational capacity. On the other hand, computational principles underlying these extraordinary feats are still only partially understood. The theoretical framework of reinforcement learning (RL) provides an ideal focal point to bring the two fields together for mutual benefit. In particular, we analyze and compare representations of space in robot and insect navigation models through the lens of RL, as the efficiency of insect navigation is likely rooted in an efficient and robust internal representation, linking retinotopic (egocentric) visual input with the geometry of the environment. While RL has long been at the core of robot navigation research, current computational theories of insect navigation are not commonly formulated within this framework, but largely as an associative learning process implemented in the insect brain, especially in the mushroom body (MB). Here we propose specific hypothetical components of the MB circuit that would enable the implementation of a certain class of relatively simple RL algorithms, capable of integrating distinct components of a navigation task, reminiscent of hierarchical RL models used in robot navigation. We discuss how current models of insect and robot navigation are exploring representations beyond classical, complete map-like representations, with spatial information being embedded in the respective latent representations to varying degrees.

How labels shape visuocortical processing in infants.

The current study examined the extent to which labels shape visuocortical processing during the first year of life during a brief (~6-min) associative learning task. Images of computer-generated artificial objects were paired with either individual-level (e.g., Jimmy, Boris) or category-level labels (e.g., Hitchel) while event-related potentials were recorded in response to the onset of the visual stimulus in 6- (n=41), 9- (n=27), and 12-month-old (n=28) infants. Analyses examined experience-dependent visuocortical changes within and across trials, label conditions, and ages. Overall, results demonstrate that infants deploy greater visuocortical resources during the first half of associative learning trials and to stimuli paired with category-level relative to individual-level labels. Waveform morphologies also differed between stimuli paired with individual- and category-level labels and across the age groups, with more complex deflections and amplitude differences between label type at 9- and 12-month-olds, but not 6-month-old infants. The present results highlight the importance of associative learning during infancy and suggest that category- versus individual-level labels differentially direct infant attention and visuocortical processing.

Analysis of fast calcium dynamics of honey bee olfactory coding

Odour processing exhibits multiple parallels between vertebrate and invertebrate olfactory systems. Insects, in particular, have emerged as relevant models for olfactory studies because of the tractability of their olfactory circuits. Here, we used fast calcium imaging to track the activity of projection neurons in the honey bee antennal lobe (AL) during olfactory stimulation at high temporal resolution. We observed a heterogeneity of response profiles and an abundance of inhibitory activities, resulting in various response latencies and stimulus-specific post-odour neural signatures. Recorded calcium signals were fed to a mushroom body (MB) model constructed implementing the fundamental features of connectivity between olfactory projection neurons, Kenyon cells (KC), and MB output neurons (MBON). The model accounts for the increase of odorant discrimination in the MB compared to the AL and reveals the recruitment of two distinct KC populations that represent odorants and their aftersmell as two separate but temporally coherent neural objects. Finally, we showed that the learning-induced modulation of KC-to-MBON synapses can explain both the variations in associative learning scores across different conditioning protocols used in bees and the bees' response latency. Thus, it provides a simple explanation of how the time contingency between the stimulus and the reward can be encoded without the need for time tracking. This study broadens our understanding of olfactory coding and learning in honey bees. It demonstrates that a model based on simple MB connectivity rules and fed with real physiological data can explain fundamental aspects of odour processing and associative learning.

Chronic social defeat stress gives rise to social avoidance through fear learning

Chronic social defeat stress (CSDS), a widely used rodent model of stress, reliably leads to decreased social interaction in stress susceptible animals. Here, we investigate a role for fear learning in this response using male 129 Sv/Ev mice, a strain that is more vulnerable to CSDS than the commonly used C57BL/6 strain. We first demonstrate that defeated 129 Sv/Ev mice avoid a CD-1 mouse, but not a conspecific, indicating that motivation to socialize is intact in this strain. CD-1 avoidance is characterized by approach behavior that results in running in the opposite direction, activity that is consistent with a threat response. We next test whether CD-1 avoidance is subject to the same behavioral changes found in traditional models of Pavlovian fear conditioning. We find that associative learning occurs across 10 days CSDS, with defeated mice learning to associate the color of the CD-1 coat with threat. This leads to the gradual acquisition of avoidance behavior, a conditioned response that can be extinguished with 7 days of repeated social interaction testing (5 tests/day). Pairing a CD-1 with a tone leads to second-order conditioning, resulting in avoidance of an enclosure without a social target. Finally, we show that social interaction with a conspecific is a highly variable response in defeated mice that may reflect individual differences in generalization of fear to other social targets. Our data indicate that fear conditioning to a social target is a key component of CSDS, implicating the involvement of fear circuits in social avoidance.

The mesolimbic system and the loss of higher order network features in schizophrenia when learning without reward.

Schizophrenia is characterized by a loss of network features between cognition and reward sub-circuits (notably involving the mesolimbic system), and this loss may explain deficits in learning and cognition. Learning in schizophrenia has typically been studied with tasks that include reward related contingencies, but recent theoretical models have argued that a loss of network features should be seen even when learning without reward. We tested this model using a learning paradigm that required participants to learn without reward or feedback. We used a novel method for capturing higher order network features, to demonstrate that the mesolimbic system is heavily implicated in the loss of network features in schizophrenia, even when learning without reward. fMRI data (Siemens Verio 3T) were acquired in a group of schizophrenia patients and controls (n=78; 46 SCZ, 18 ≤ Age ≤ 50) while participants engaged in associative learning without reward-related contingencies. The task was divided into task-active conditions for encoding (of associations) and cued-retrieval (where the cue was to be used to retrieve the associated memoranda). No feedback was provided during retrieval. From the fMRI time series data, network features were defined as follows: First, for each condition of the task, we estimated 2nd order undirected functional connectivity for each participant (uFC, based on zero lag correlations between all pairs of regions). These conventional 2nd order features represent the task/condition evoked synchronization of activity between pairs of brain regions. Next, in each of the patient and control groups, the statistical relationship between all possible pairs of 2nd order features were computed. These higher order features represent the consistency between all possible pairs of 2nd order features in that group and embed within them the contributions of individual regions to such group structure. From the identified inter-group differences (SCZ ≠ HC) in higher order features, we quantified the respective contributions of individual brain regions. Two principal effects emerged: 1) SCZ were characterized by a massive loss of higher order features during multiple task conditions (encoding and retrieval of associations). 2) Nodes in the mesolimbic system were over-represented in the loss of higher order features in SCZ, and notably so during retrieval. Our analytical goals were linked to a recent circuit-based integrative model which argued that synergy between learning and reward circuits is lost in schizophrenia. The model's notable prediction was that such a loss would be observed even when patients learned without reward. Our results provide substantial support for these predictions where we observed a loss of network features between the brain's sub-circuits for a) learning (including the hippocampus and prefrontal cortex) and b) reward processing (specifically constituents of the mesolimbic system that included the ventral tegmental area and the nucleus accumbens. Our findings motivate a renewed appraisal of the relationship between reward and cognition in schizophrenia and we discuss their relevance for putative behavioral interventions.