Primates chunk simultaneously-presented memoranda

Abstract

Though much research has characterized both the behavior and electrophysiology of spatial memory for single targets in non-human primates, we know much less about how multiple memoranda are handled. Multiple memoranda may interact in the brain, affecting the underlying representations. Mnemonic resources are famously limited, so items may compete for "space" in memory or may be encoded cooperatively or in a combined fashion. Understanding the mode of interaction will inform future neural studies. As a first step, we quantified interactions during a multi-item spatial memory task. Two monkeys were shown 1-4 target locations. After a delay, the targets reappeared with a novel target and the animal was rewarded for fixating the novel target. Targets could appear either all at once (simultaneous) or with intervening delays (sequential). We quantified the degree of interaction with memory rate correlations. We found that simultaneously presented targets were stored cooperatively while sequentially presented targets were stored independently. These findings demonstrate how interaction between concurrently memorized items depends on task context. Future studies of multi-item memory would be served by designing experiments to either control or measure the mode of this interaction.

Keywords: chunking; monkey; multi-item; primate; working memory.

FIGURE 1

Venn diagrams of pairs of random variables. (A) Two random variables that are statistically independent. (B) Two random variables with positive correlation. (C) Two random variables with negative correlation. In this case, the variables are mutually exclusive.

FIGURE 2

Single- and multi-item spatial memory tasks. Trials begin with a 700 ms fixation of a central point. For n-item simultaneous trials, n peripheral targets appear 13 degrees of visual angle (dva) out from the fixation point. Animals continue to fixate through a variable delay period. For n-item sequential trials, a single target appears first and subsequent targets appear throughout the full delay. For all trial types, all of the previously shown peripheral targets reappeared at the end of the delay along with a single novel target. Animals received liquid rewards for making a saccade to the novel target. Delay period lengths were varied across sessions and ranged from 2 to 8 s.

FIGURE 3

Task performance. Hit rates for each trial type (simultaneous or sequential) as a function of memory load. Error bars depict 95% confidence intervals. Dotted lines indicate chance levels.

FIGURE 4

Target choice probabilities and memory probabilities during the 2-item simultaneous task. (A) The trial-wise probabilities of the animal choosing either the novel target or a memory target (either target 1 or 2) given that the animal chose a target (i.e., the animal moved his eyes to within 4.5 dva of either the novel target or a memory target). The animals rarely failed to choose a target after leaving the fixation point (4% of trials, see Supplementary Figure 4). (B) We assume a model in which the probabilities of choosing targets depend on the probabilities of remembering the targets (see section “Materials and methods”). We estimated a range of solutions, varying an assumed value for the probability of forgetting both targets. Here, we show two extreme solutions—when the probability of forgetting both targets is minimal (left), i.e., zero, and when the probability of forgetting both targets is maximal (right). R, remembered; F, forgotten. Note that these values do not map in a one-to-one fashion with the values in panel (A). (C) Estimated correlation values between the memory of the two targets. Gray segments indicate the range-estimates of possible correlation coefficient values estimated with standard trials. With the addition of an additional trial type (omission trials, see section “Results”), we are able to make point estimates of correlation. The black circle and line indicate a point-estimate mean and 95% confidence interval, respectively.

FIGURE 5

Target choice probabilities and memory probabilities during the 2-item sequential task. Format identical to Figure 4.