Abstract
An informative cue that directs attention to one of several items in working memory improves subsequent recall of that item. Here we examine the mechanism of this retro-cue effect using a model of short-term memory based on neural population coding. Our model describes recalled feature values as the output of an optimal decoding of spikes generated by a tuned population of neurons. This neural model provides a better account of human recall data than an influential model that assumes errors can be described as a mixture of normally distributed noise and random guesses. The retro-cue benefit is revealed to be consistent with a higher firing rate of the population encoding the cued versus uncued items, with no difference in tuning specificity. Additionally, a retro-cued item is less likely to be swapped with another item in memory, an effect that can also be explained by greater activity of the underlying population. These results provide a parsimonious account of the effects of retrospective attention on recall and demonstrate a principled method for investigating neural representations with behavioral tasks.
Highlights
Recall of sensory information from short-term memory storage is imperfect; not all memory representations are alike: cueing paradigms have revealed that memory resources can be allocated under voluntary control
We found that behavioral data drawn from existing studies were consistent with a relatively enhanced firing rate of the population encoding a cued item, whereas tuning width remained unchanged between cued and uncued items
These observations parallel neurophysiological results related to prospective orienting of attention to visual stimuli, where it has been shown that directing attention to a cued stimulus scales the firing rate of sensory neurons without a change in tuning width (McAdams & Maunsell, 1999; Motter, 1993)
Summary
Recall of sensory information from short-term memory storage is imperfect; not all memory representations are alike: cueing paradigms have revealed that memory resources can be allocated under voluntary control. Benefits are observed for items cued at the time of presentation (Gorgoraptis, Catalao, Bays, & Husain, 2011; Sperling, 1960), and for items cued retrospectively (Griffin & Nobre, 2003; Landman, Spekreijse, & Lamme, 2003; Pertzov, Bays, Joseph, & Husain, 2013; Souza, Rerko, Lin, & Oberauer, 2014). In such retro-cueing experiments, one item from a visual memory array is indicated during the retention interval by, e.g., a spatial cue. Proposals include the removal from memory of redundant information related to uncued items (Souza & Oberauer, 2016), a strengthening of the cued item’s memory representation over and above its original encoding strength (Rerko, Souza, & Oberauer, 2014; Souza et al, 2014; Souza, Rerko, & Oberauer, 2015), protection of the cued item from temporal decay or interference from other items (Pertzov et al, 2013; Pertzov, Manohar, & Husain, 2016), and protection from interference that arises from subsequent visual input (Makovski, Watson, Koutstaal, & Jiang, 2010; Souza & Oberauer, 2016; Souza, Rerko, & Oberauer, 2016)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
