Erdem Pulcu

Ketamine attenuates habenula activity in response to aversive outcomes during Pavlovian learning

This study shows ketamine reduces habenula activity during aversive learning in humans, supporting preclinical findings on its antidepressant mechanisms.

Valence-specific representation of uncertainty in the anterior cingulate is impaired in those with affective symptoms

The paper shows that individuals with anxiety and depression have impaired ability to adapt learning rates to reward and punishment volatility, linked to reduced neural encoding in the anterior cingulate cortex.

A nonlinear relationship between prediction errors and learning rates in human reinforcement learning

The paper demonstrates a nonlinear relationship between prediction errors and learning rates in human reinforcement learning, showing it can be modeled by an exponential-logarithmic function.

Humans adapt rationally to approximate estimates of uncertainty

This study shows humans adapt to volatility but not noise, with imprecise noise estimates leading to misattribution as volatility.

Value-based decision-making between affective memories

The paper investigates how affective memories influence value-based decision-making using reinforcement learning models and shows a positive bias in recall preferences.

Affective biases encoded by the central arousal systems dynamically modulate inequality aversion in human interpersonal negotiations

This study shows that affective biases from central arousal systems modulate inequality aversion in human negotiations, with pupil-linked activity encoding proposer reactions that influence decision-making.

Value computations underlying human proposer behaviour in the Ultimatum Game

This study models human proposer behavior in the Ultimatum Game, finding that risk perception is common across decisions, while nonlinear weighting of acceptance probabilities is unique to social interactions involving inferred opponent valuations.

The effects of the angiotensin II receptor antagonist losartan on appetitive versus aversive learning

This study investigates how losartan affects appetitive versus aversive learning in humans using a reinforcement learning task, finding it enhances learning from positive outcomes more than negative ones.

Affective Bias as a Rational Response to the Statistics of Rewards and Punishments

The paper shows that affective biases in processing rewards and punishments reflect estimates of their information content, with implications for depression treatment.