A Cerebellar Framework for Predictive Coding and H

Authors: Dennis J L G Schutter

Publish Date: 2015/08/07

Volume: 15, Issue: 1, Pages: 30-33

PDF Link

Abstract

Depressive disorder is associated with abnormalities in the processing of reward and punishment signals and disturbances in homeostatic regulation These abnormalities are proposed to impair error minimization routines for reducing uncertainty Several lines of research point towards a role of the cerebellum in reward and punishmentrelated predictive coding and homeostatic regulatory function in depressive disorder Available functional and anatomical evidence suggests that in addition to the corticolimbic networks the cerebellum is part of the dysfunctional brain circuit in depressive disorder as well It is proposed that impaired cerebellar function contributes to abnormalities in predictive coding and homeostatic dysregulation in depressive disorder Further research on the role of the cerebellum in depressive disorder may further extend our knowledge on the functional and neural mechanisms of depressive disorder and development of novel antidepressant treatments strategies targeting the cerebellumPsychological wellbeing to a significant extend depends on the individual’s ability to deal with uncertainty 1 Experience of uncertainty is inversely related to experiential feelings of control and results from actions that do not lead to the anticipated outcome The mismatch between anticipated and actual outcome gives rise to a prediction error signal generated by the brain which is indicative for a disruption of internal bodily homeostasis 2 Prediction error signals typically lead to a cascade of physiological and psychological processes that serve to reestablish equilibrium 3 The idea of reducing uncertainty by optimizing predictability is known as the error minimization routine of predictive coding 2 The error minimization routine is considered to form the basis of the organism’s ability to construct and update internal models that allow for successful adaptation under changing conditions 2 Feedbackrelated reward and punishment signals arguably play an important role in the formation of internal prediction models and in shaping context appropriate behavior 4 The fact that depressive disorder is characterized by hypersensitivity to punishment and hyposensitivity to reward suggests that in patients reward and punishment signals provide suboptimal input for the error minimization routine to work properly Support for this idea comes from findings showing that patients suffering from depression show abnormal neural responses to unexpected outcomes as well as context updating in response to feedback 5 Atypical physiological response patterns to stress and abnormal biorhythms in patients with mood disorders further hint at a dysregulation of the neural circuitry concerned with bodily homeostasis The subjective experience of negative mood and lack of control may thus represent a phenomenological proxy for problems in homeostatic regulatory function 6Major depressive disorder is associated with volume reductions in the frontal cortex hippocampus and striatum 21 In addition abnormal patterns of activity in the frontal cortex striatum globus pallidus amygdala and cerebellum have been observed 15 These metaanalytic results support the view that depressive disorder involves abnormalities in the rewardpunishmentrelated limbic circuits paralleled by dysfunctional brain structures implicated in topdown regulation Further evidence for compromised regulation capacity and topdown regulation in depressive disorder was demonstrated in a recent metaanalysis which found lower functional connectivity between frontal and limbic structures 22 The same study also showed that depressed patients had lower connectivity between parietal and frontoparietal regions involved in attending to the external environment but hyperconnectivity in the default network associated with internal selfreferential thoughts and feelings 22 Furthermore the cerebellum was hypoconnected to the parietal cortex a finding which concurs with the proposed cerebellar role in reality monitoring and predictive codingIn agreement with conventional motor theories that conceptualize the cerebellum as a learning machine three computational primitives for understanding the relations between homeostasis predictive coding and depressive disorder are important 10 14 Firstly an internal forward model that is able to predict/anticipate the consequences of behavior Secondly an internal feedback model that is able to compare and detect mismatches between predicted and actual outcomes Thirdly an error minimization routine that actively modifies the internal forward and feedback models enabling more accurate predictions of the environment From this viewpoint behavioral adaptive responses involve the cerebellum performing Bayesian probability modeling that uses reward and punishment signals as inputs to update the priors to minimize uncertainty and regain bodily homeostasis 2 Results from functional neuroimaging studies show that cerebellar activity correlates with error monitoring and probabilistic inferences in decisionmaking and context updating 23According to the uniform cerebellar transform function the homogenous microstructure of the cerebellum allows for the processing of multimodal input signals originating from the rich cerebellar connections with cortical and limbic parts of the brain 9 Anxiety and depression can be viewed as phenomenological manifestations of disrupted bodily homeostasis and uncertainty that prompts the cerebellum to update the priors of the internal model to minimize prediction errors The conceptual framework predicts that in line with the universal cerebellar transform function neuroticism and mood disorders are associated with problems in updating the internal model Problems with updating the priors of the internal model will affect the prediction error minimization routine and contribute to feelings of uncertainty and loss of control This view builds upon the central idea that the cerebellum is important for synchronizing cortical cognitive and limbic motivational information processing streams to fit contextual demands In addition the latter also gives a possible mechanistic account for why abnormalities in the cerebellar transform function could result in disorganized thought and feelings 9 Abnormalities in cerebellar predictive coding may offer a theoretical framework to explain at least in part why anxiety and depression are associated with subjective reports of experiencing loss of control and feelings of helplessness Finally the present framework may provide a starting point for developing novel noninvasive brain stimulation protocols for the treatment of depressive disorder by targeting the cerebellum 18In conclusion research has been discussed in support of the idea that the cerebellum contributes to reward and punishmentrelated predictive coding and plays a role in the regulation of bodily homeostasis which is proposed to be dysfunctional in depressive disorderOpen Access This article is distributed under the terms of the Creative Commons Attribution 40 International License http//creativecommonsorg/licenses/by/40/ which permits unrestricted use distribution and reproduction in any medium provided you give appropriate credit to the original authors and the source provide a link to the Creative Commons license and indicate if changes were made

Keywords: