A dynamic code for economic object valuation in prefrontal cortex neurons

Abstract

Neuronal reward valuations provide the physiological basis for economic behaviour. Yet, how such valuations are converted to economic decisions remains unclear. Here we show that the dorsolateral prefrontal cortex (DLPFC) implements a flexible value code based on object-specific valuations by single neurons. As monkeys perform a reward-based foraging task, individual DLPFC neurons signal the value of specific choice objects derived from recent experience. These neuronal object values satisfy principles of competitive choice mechanisms, track performance fluctuations and follow predictions of a classical behavioural model (Herrnstein’s matching law). Individual neurons dynamically encode both, the updating of object values from recently experienced rewards, and their subsequent conversion to object choices during decision-making. Decoding from unselected populations enables a read-out of motivational and decision variables not emphasized by individual neurons. These findings suggest a dynamic single-neuron and population value code in DLPFC that advances from reward experiences to economic object values and future choices.

Experimentally revealed stochastic preferences for multi-component choice options

“…We assessed the contribution of individual IPs to the hyperbolically fitted ICs with three tests. (1) Using a leave-one-out analysis, we compared ICs fitted to all five IPs with ICs fitted with one IP left out and found good correspondence in all of four tests ( Figure S2). (2) Using a previously developed single-dimensional linear support vector machine (SVM) algorithm (Grabenhorst, Hernadi, & Schultz, 2016;Tsutsui, Grabenhorst, Kobayashi, & Schultz, 2016) and (3) a two-dimensional linear discriminant analysis (LDA), we assessed the accuracy of reversely assigning individual IPs to their original preference level. Both decoders reported across-IC accuracies largely in the 70% -100% range, and the LDA showed only random distinction along-ICs ( Figure S3; Table S3 left); as a control, shuffled data did not discriminate between preference levels (SVM accuracies of 44.7% -54.6%; Table S4 left).…”

“…Our main test employed a binary support vector machine (SVM) decoder separately on each individual participant. We used similar methods as previously described for predicting choice from neuronal activity (Tsutsui, Grabenhorst, Kobayashi, & Schultz, 2016). The SVM algorithm considered 5 IP bundles from each of 2 revealed preference levels (total of 10 IPs that had been assessed 12 times at each position in each participant) ( Figure S3A).…”

“…For assessing chance decoding, we shuffled the 2 x 5 matrix. Our earlier work has shown that increasing the number of analysis trials from 10 to 20 resulted in similar accuracy (Grabenhorst, Hernadi, & Schultz, 2016;Tsutsui, Grabenhorst, Kobayashi, & Schultz, 2016). The SVM was implemented with custom written software in Matlab R2015b (Mathworks) using the functions svmtrain and svmclassify with linear kernel (our previous work had shown that use of nonlinear kernels did not improve decoder performance; Tsutsui, Grabenhorst, Kobayashi, & Schultz, 2016).…”

Comparative Overview of Visuospatial Working Memory in Monkeys and Rats

“…Much evidence from human neuropsychological and neuroimaging studies indicate that the human PFC is involved not only in visuospatial working memory but also in nonspatial working memory of various modalities, as well as many other aspects of cognitive and executive control functions (e.g., Owen et al, 1996; Koechlin et al, 1999; Olesen et al, 2004; for review, Stuss and Knight, 2002; Fuster, 2008; Passingham and Wise, 2012). Monkey electrophysiological studies have shown the neural correlates of various cognitive functions besides working memory within the PFC on the single-neuron level, such as response inhibition (Watanabe, 1986b), attentional control (Sakagami and Tsutsui, 1999; Lebedev et al, 2004), categorical recognition (Freedman et al, 2001; Antzoulatos and Miller, 2011; Tsutsui et al, 2016b), numerical recognition (Nieder et al, 2002), rule-based judgments (Wallis et al, 2001; Mansouri et al, 2006; Yamada et al, 2010), value-based decision making (Barraclough et al, 2004; Cai and Padoa-Schioppa, 2014; Tsutsui et al, 2016a), and complex action planning (Mushiake et al, 2006). We have no intention to insist that the function of the entire PFC can be solely explained by working memory, and indeed we admit that even the above mentioned list of PFC functions is not at all exhaustive.…”

Interacting roles of lateral and medial Orbitofrontal cortex in decision-making and learning : A system-level computational model

“…The 182 output activities of mOFC are fed into its CBG loop. It has been shown that one of 183 the general function of populations in the PFC is to maintain history of decision 184 events such as previous action, previous reward etc [68]. Accordingly, we implemented 185 a simple history of rewards in mOFC, without cue-specific information.…”