Hippocampal BOLD response during category learning predicts subsequent performance on transfer generalization

Abstract

To test a prediction of our previous computational model of cortico-hippocampal interaction (Gluck and Myers [1993, 2001]) for characterizing individual differences in category learning, we studied young healthy subjects using an fMRI-adapted category-learning task that has two phases, an initial phase in which associations are learned through trial-and-error feedback followed by a generalization phase in which previously learned rules can be applied to novel associations (Myers et al. [2003]). As expected by our model, we found a negative correlation between learning-related hippocampal responses and accuracy during transfer, demonstrating that hippocampal adaptation during learning is associated with better behavioral scores during transfer generalization. In addition, we found an inverse relationship between Blood Oxygenation Level Dependent (BOLD) activity in the striatum and that in the hippocampal formation and the orbitofrontal cortex during the initial learning phase. Conversely, activity in the dorsolateral prefrontal cortex, orbitofrontal cortex and parietal lobes dominated over that of the hippocampal formation during the generalization phase. These findings provide evidence in support of theories of the neural substrates of category learning which argue that the hippocampal region plays a critical role during learning for appropriately encoding and representing newly learned information so that that this learning can be successfully applied and generalized to subsequent novel task demands.

Keywords: BOLD activity; basal ganglia; category learning; computational model; cortico-hippocampal interaction; transfer generalization.

Figure 1

(A) Screen events on a sample trial of phase 1 (concurrent discrimination) of the two‐phase learning and generalization task. On each trial, the object pair is presented in random left‐right order and a prompt appears. If the participant responds correctly, the chosen object is raised to reveal a smiley face icon underneath. In this pair, the color of the objects differs but shape is the same and therefore irrelevant; in other pairs (not shown), the shape of the objects differs but color is the same (irrelevant). (B) Screen events on a sample trial of phase 2 (transfer generalization): events are similar to phase 1, but the objects are changed so that the relevant dimension (here, the color) is the same, whereas the irrelevant dimension (here, the shape) is novel. [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]

Figure 2

Stimulus set used for concurrent discrimination (phase 1, left panels) and transfer (phase 2, right panels). Each pair of objects differed either by color (top panels) or by shape (bottom panels). For transfer (phase 2, right panels), the relevant dimension stayed the same, while the irrelevant dimension was changed. [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]

Figure 3

Behavioral performance during each stage of the task. RT, reaction times.

Figure 4

Blood Oxygenated Level Dependent (BOLD) activity (A.U., arbitrary units) during each task stage (early and late learning, no‐feedback and transfer generalization) in different regions of interest. Bar plots represent mean values ± standard errors. Color bars represent F statistics. Coordinates (X, Y, Z) are given in the Montreal Neurological Institute (MNI) space. [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]

Figure 5

Subject‐specific Blood Oxygenated Level Dependent (BOLD) activity in the bilateral putamen (for the contrast late>early learning) positively correlates with individual differences in behavioral accuracy during learning (defined as the difference between correct responses during the late task stage minus correct responses during the early task stage). Black lines represent regression lines while red lines represent the 95% confidence interval. Color bars represent T statistics. FWE, svc, Family Wise Error, small volume correction. Coordinates (X, Y, Z) are given in the Montreal Neurological Institute (MM) space. [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]

Figure 6

Subject‐specific Blood Oxygenated Level Dependent (BOLD) activity in the bilateral hippocampus (for the contrast transfer>no‐feedback) negatively correlates with individual differences in behavioral accuracy during testing (defined as the difference between correct responses during the transfer task stage minus correct responses during the no‐feedback task stage). Black lines represent regression lines while red lines represent the 95% confidence intervals. Color bars represent T statistics. FWE, svc, Family Wise Error, small volume correction. Coordinates (X, Y, Z) are given in the Montreal Neurological Institute (MNI) space. [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]

Figure 7

Subject‐specific Blood Oxygenated Level Dependent (BOLD) activity in the bilateral Hippocampus (for the contrast late>early learning) negatively correlates with individual differences in behavioral accuracy during transfer (defined as the difference between correct responses during the transfer task stage minus correct responses during the no‐feedback task stage). Black lines represent regression lines while red lines represent the 95% confidence intervals. Color bars represent T statistics. FWE, svc, Family Wise Error, small volume correction. Coordinates (X, Y, Z) are given in the Montreal Neurological Institute (MNI) space. [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]