Does Signed Prediction Error drive Declarative Memory? Evidence from Variable Choice Paradigms

Prediction error (PE) is the degree of conflict between predictions and new information. For a binary reward outcome, PE may be signed (positive if the outcome was correctly predicted and negative if the outcome was incorrectly predicted), or unsigned (absolute value of “surprise”). Using a “variable choice” paradigm, De Loof et al. (2018) examined the role of PE in one-shot learning of unknown translations of known words, showing associative memory for the translation to be greater when (financial) reward was more unexpected, and lesser when an expected reward was not received (i.e., signed PE); an effect they replicated in several subsequent studies. However, previous work on PE in declarative memory has assumed that memory is greater when an outcome is more unexpected (i.e., unsigned PE). We replicated De Loof et al.’s paradigm with and without financial reward, and found that memory was explained slightly better by unsigned PE (Experiment 1A-1B). However, we then identified a potential confound in the paradigm which could explain results without any role of PE. Simulations confirmed this possibility, and we designed and tested a modified version of the paradigm (Experiment 2). Results were inconsistent with the PE account and suggested that findings from variable choice paradigms might be an artefact of the experimental design. Overall, we conclude that variable choice paradigms may not be well suited to investigate the role of PE in one-shot declarative learning, and that the purported role of signed PE in declarative memory requires further investigation.