A Data-Driven Characterization of the Adaptability of Human Behavior
Open AccessA central aspect of human behavior is that people are constantly extracting regularities from their environment, and using those regularities to guide subsequent behavior in order to be more efficient. While in many cases this adaptability is advantageous, the automaticity of the process can lead individuals to adjust their behavior based on perceived regularities that have no ground truth (i.e., the events in the real world are independent of each other). However, an important open question is how people aggregate information across prior events to guide behavior. Recent proposals suggest that the mechanism by which information accumulates to promote behavioral change may be supported by simple Hebbian learning (e.g., Reber, 2013), yielding several behavioral predictions: behavioral adaptation should be 1) directly proportional to experience, 2) domain-general, and 3) maintained across delays consistent with known mechanisms of synaptic change. Across three studies, the moment-to-moment adaptation of human behavior was characterized and each of these predictions were verified. In the first study, the relationship between prior events and subsequent behavior was established, such that the adaptation of behavior was directly proportional to the statistics of prior events and was explained by the same function across two tasks and multiple task features (consistent with domain-generality). In the second study, behavioral changes were examined with more specificity, such that the manner in which two underlying processes of behavior (e.g., motor initiation and decision processes) differentially adapted to prior events was explored. Lastly, the third study investigated the maintenance of aggregated information across delays of varying lengths, verifying predictions based on known short- and long-term mechanisms of synaptic weight change. Taken together, the findings of the three studies provide a comprehensive characterization of the adaptation of human behavior that fits into a mechanistic framework of distributed cortical plasticity, with important experimental and translational implications.
- All rights reserved
Notice to Authors
If you are the author of this work and you have any questions about the information on this page, please use the Contact form to get in touch with us.