Impacts and interactions of stress, noradrenaline and serotonin signalling on probabilistic reversal learning
Stupart, O.; Wilod Versprille, L. J. F.; Zuhlsdorff, K.; Velazquez-Sanchez, C.; Bailey, M. C. D.; Chen, J.; Lawson, R. P.; Dalley, J. W.
Show abstract
Rationale: Early life stress (ELS) is acknowledged to underlie cognitive and emotional abnormalities linked to stress-related mood disorders. ELS can lead to persistent biases in how uncertain feedback is processed to affect the flexibility of decision-making. Objectives: (1) To investigate the effects of ELS on the flexibility of rats trained on a serial probabilistic reversal learning (PRL) task involving spurious positive and negative feedback. (2) To elucidate the involvement of the stress hormone corticosterone and the noradrenergic and serotonergic systems in modulating how ELS affects PRL. Methods: Male and female rats were intermittently separated from maternal care on postnatal days five to nineteen, inclusively. As adults, the same rats were trained on a deterministic reversal learning task involving certain rewarded or non-rewarded outcomes followed by a PRL task where correct and incorrect responses were rewarded on 80% and 20% of trials, respectively. Dose-dependent effects of the beta-blocker, propranolol, selective serotonin reuptake inhibitor, citalopram and corticosterone were subsequently determined. Results: ELS resulted in an increased responsivity to feedback, specifically in males making more win-stay responses following a reward, that was associated with an increased punishment learning rate. In both control and MS rats, propranolol increased feedback sensitivity, but delayed updating following a rule switch. In contrast, neither citalopram nor corticosterone significantly affected reversal learning. Conclusions: ELS is sufficient to cause persistent changes in how feedback is processed by male rats on a reversal learning task. Activation of beta-adrenergic receptors may be necessary for updating learned associations during decision-making involving uncertain feedback.
Matching journals
The top 7 journals account for 50% of the predicted probability mass.