Expectation-driven changes in cortical functional connectivity influence working memory and long-term memory performance

Abstract

Expectations generated by predictive cues increase the efficiency of perceptual processing for complex stimuli (e.g., faces, scenes); however, the impact this has on working memory (WM) and long-term memory (LTM) has not yet been investigated. Here, healthy young adults performed delayed-recognition tasks that differed only in stimulus category expectations, while behavioral and functional magnetic resonance imaging data were collected. Univariate and functional-connectivity analyses were used to examine expectation-driven, prestimulus neural modulation, networks that regulate this modulation, and subsequent memory performance. Results revealed that predictive category cueing was associated with both enhanced WM and LTM for faces, as well as baseline activity shifts in a face-selective region of the visual association cortex [i.e., fusiform face area (FFA)]. In addition, the degree of functional connectivity between FFA and right inferior frontal junction (IFJ), middle frontal gyrus (MFG), inferior frontal gyrus, and intraparietal sulcus correlated with the magnitude of prestimulus activity modulation in the FFA. In an opposing manner, prestimulus connectivity between FFA and posterior cingulate cortex, a region of the default network, negatively correlated with FFA activity modulation. Moreover, whereas FFA connectivity with IFJ and the precuneus predicted enhanced expectation-related WM performance, FFA connectivity with MFG predicted LTM improvements. These findings suggest a model of expectancy-mediated neural biasing, in which a single node (e.g., FFA) can be dynamically linked or disconnected from different brain regions depending on prestimulus expectations, and the strength of distinct connections is associated with WM or LTM benefits.

Figure 1.

Experimental paradigm. All participants performed four tasks (SKf trials, SKs trials, SUf and SUs trials, and PVf and PVs trials), which were blocked and counterbalanced and for which stimuli were randomized. For passive-view trials, delay and probe periods were removed (see Materials and Methods). Note that fixation crosses were green, red, and gray for expectation, delay, and intertrial interval (ITI) periods, respectively (not shown in figure).

Figure 2.

Behavioral performance. A, WM accuracy. Compared with the neutral conditions, specific predictive foreknowledge resulted in increased WM accuracy for face trials (*p < 0.05), but not scene trials (p > 0.05). B, LTM indices. Compared with the neutral conditions, specific predictive foreknowledge was associated with higher LTM recognition scores for face trials (*p < 0.05), but not scene trials (p > 0.05). Error bars indicate SEM.

Figure 3.

FFA/PPA univariate activity: expectation-related baseline shifts. For the FFA, expectation-related activity modulation (i.e., baseline shift) was greater for the predictive condition than for neutral and passive-view face trials, as well as for remember scene trials (faces were uncued) (all comparisons, *p < 0.05). The PPA did not display significant expectation-related activity (all comparisons, p > 0.05). Error bars indicate SEM.

Figure 4.

FFA functional connectivity: main effects. A, Axial and sagittal slices illustrate FFA-seed functional connectivity main effects. This analysis revealed that regions in the left MFG (1), right MFG (2), right IFJ (3), right IPS (4), and precuneus (5) (red arrows) were functionally connected with the FFA during the expectation period in the face-predictive condition (SKf). B, Neutral condition (SUf) trials did not reveal FFA connectivity with the regions in A. C, For passive-view face trials (PVf), regions belonging to the default network, including PCC (7) and mPFC (8) (red arrows), were significantly functionally connected with the FFA during the expectation period.

Figure 5.

FFA functional connectivity: contrasts. Expectation-related FFA connectivity contrasts of SKf > SUf (A) and PVf > (SKf + SUf) (B). The numbers correspond to regions identified in the main effects analysis (Fig. 4), which remained significant in the contrast, as well as right IFG (6). Stereotaxic MNI coordinates and mean p values for significant regions are shown in Table 1.

Figure 6.

Regression analysis: functional connectivity/FFA activity. The right IFJ (A), right MFG (B), right IFG (C), and right IPS (D) display significant positive correlations between FFA connectivity and univariate FFA activity modulation during face-specific expectation periods in an across-participant regression analysis. In contrast, the PCC (E) displays a negative correlation between task-differential FFA connectivity (SKf–PVf) and univariate FFA activity modulation (SKf–PVf) during the expectation periods. ROIs localized on three-dimensional brain images and task-specific FFA connectivity measures are displayed in the insets above the scatterplots (*p < 0.05). Error bars indicate SEM.

Figure 7.

Regression analysis: neurobehavioral correlations. WM and LTM performance measures were correlated with connectivity between FFA and distinct frontoparietal regions during the stimulus expectation period. Brain figures illustrate the ROIs from the FFA connectivity contrast SKf > SUf whose task-differential connectivity measures (SKf–SUf) were positively correlated with task-differential behavior (WM or LTM) (see Materials and Methods). Shown in red are regions displaying expectation-related, functional connectivity with the FFA that correlates with WM performance (right IFJ, right precuneus). Shown in blue are regions displaying expectation-related functional connectivity with the FFA that correlates with LTM measures (left MFG). The corresponding scatterplots are shown below each region.