Primary and secondary rewards differentially modulate neural activity dynamics during working memory

Abstract

Background: Cognitive control and working memory processes have been found to be influenced by changes in motivational state. Nevertheless, the impact of different motivational variables on behavior and brain activity remains unclear.

Methodology/principal findings: The current study examined the impact of incentive category by varying on a within-subjects basis whether performance during a working memory task was reinforced with either secondary (monetary) or primary (liquid) rewards. The temporal dynamics of motivation-cognition interactions were investigated by employing an experimental design that enabled isolation of sustained and transient effects. Performance was dramatically and equivalently enhanced in each incentive condition, whereas neural activity dynamics differed between incentive categories. The monetary reward condition was associated with a tonic activation increase in primarily right-lateralized cognitive control regions including anterior prefrontal cortex (PFC), dorsolateral PFC, and parietal cortex. In the liquid condition, the identical regions instead showed a shift in transient activation from a reactive control pattern (primary probe-based activation) during no-incentive trials to proactive control (primary cue-based activation) during rewarded trials. Additionally, liquid-specific tonic activation increases were found in subcortical regions (amygdala, dorsal striatum, nucleus accumbens), indicating an anatomical double dissociation in the locus of sustained activation.

Conclusions/significance: These different activation patterns suggest that primary and secondary rewards may produce similar behavioral changes through distinct neural mechanisms of reinforcement. Further, our results provide new evidence for the flexibility of cognitive control, in terms of the temporal dynamics of activation.

Figure 1. Schematic display of Liquid and Money incentive trials.

Each trial started with a fixation cross (500 msec), followed by a cue indicating the type of incentive at stake (1000 msec; cues indicate high-incentive trials), a set of stimulus words (2500 msec), a delay period (3500 msec), a probe word (500 msec), a second delay (2500 msec) and a feedback phase, based on performance on the trial (2000 msec). Total duration: 12.5 seconds.

Figure 2. Trial-specific incentive effects.

Response times for no-incentive trials (NO), low incentive trials (LOW) and high incentive trials (HIGH) for the Money (MON) and Liquid (LIQ) blocks. Figure demonstrates main effect of incentive magnitude, reflected in faster response times on trials with higher incentive value.

Figure 3. Incentive category specific activation and overlap.

Regions showing selective transient incentive effects in Liquid (blue) and selective sustained incentive effects in Money (red). Regions showing a shift from transient activation during the Liquid condition to sustained activation during the Money condition (i.e., overlap regions) are shown in yellow.

Figure 4. Anatomical double dissociation in incentive category specific sustained activation.

Sustained activation selective to the Liquid condition was observed in the reward network as representatively shown for the ventral striatum (7, 0, −4), whereas the cognitive control network showed money-selective state effects (here shown for the DLPFC (35, 36, 22).

Figure 5. Flexibility in activation dynamics of cognitive control regions related to incentive magnitude and category.

A) Overlapping regions showing selective state effects in Money and selective item effects in Liquid. B) Overlapping regions showing a shift from sustained to transient activation across the Money and Liquid incentive conditions. Percent signal change average for all three overlapping regions, sustained effects showing averaged signal changes across the incentive block, and transient effects are averaged across frames 2 to 6 in high incentive trials. C) Time-courses for incentive trials and no-incentive trials within the Money and Liquid condition showing a shift in the peak of activation dynamics during the Liquid condition. On no-incentive trials, activation peaks late in the trial (around probe response), and on incentive trials, activation peaks earlier (during encoding). All are averaged for the three overlapping cognitive control regions. (MON_H: money high-incentive trials, MON_L: money low-incentive trials, MON_N: no-incentive trials during the Money block; LIQ_H: liquid high-incentive trials, LIQ_L: liquid low-incentive trials, LIQ_N: no-incentive trials during the Liquid block).

Figure 6. Correlations between sustained and event related BOLD responses.

Percent signal changes are averaged for the overlapping regions. Sustained and event related activations showed significant correlations in both the Liquid condition (A) and the Money condition (B).

Figure 7. Activation-performance relationship in the dorsal striatum.

Correlation between sustained activation increase and performance improvement relative to baseline observed in the Liquid condition (A) within a region in the left dorsal striatum (−9, 12, 3; panel B). During the Money condition, no such correlation was present (C). Activation effect is percent change signal increase in incentive condition relative to Baseline. Performance effect is response time improvement (in msec) during incentive condition relative to Baseline.