Influence of temporal information gaps on decision making: describing the dynamics of working memory

Abstract

Decision making involves evaluating options and predicting their likely outcomes. Traditional laboratory studies of decision making often employ tasks involving the discrimination of perceptual evidence, where sensory information is constant and presented continuously. However, during natural behavior, decision making usually involves intermittent information streams, punctuated by periods with no input. To investigate decision making under such conditions, we designed a perceptual task where participants observed tokens sequentially jumping from a central circle to one of two peripheral targets, disappearing shortly after. Participants were required to report which target they believed would have received most tokens by the trials end. Half of the trials included a temporal gap, during which no information was displayed. To better understand decision-making dynamics, we introduced specific patterns of token jumps. We found that participants made choices with less available information and disproportionally weighted the information presented immediately after the gap more heavily than they did when no gap was present. Traditional computational models, which assume uniform or gradual decay or increase of weighting of information over time, could not account for this effect. A control task with randomly presented information further confirmed that the disproportionate weighting of post-gap information is a robust feature of the decision-making process. These findings highlight the importance of studying decision making in environments with intermittent information and temporal gaps, where the timing and structure of inputs critically shape behavior. Significance statementDecision making in real-world environments often involves intermittent information, where key evidence might be separated by temporal gaps. Our study demonstrate that people disproportionately rely on information received immediately after such gaps, weighting the same information more heavily than when no gap is present. Traditional decision-making models do not account for this effect, as they lack mechanisms to capture how temporal disruptions shape evidence weighting. By understanding how gaps influence decision making, this research provides insights into how people adapt their decisions in fluctuating environments, with implications for fields ranging from neuroscience to artificial intelligence.