Prefrontal activation deficits during episodic memory in schizophrenia

Abstract

Objective: Episodic memory impairments represent a core deficit in schizophrenia that severely limits patients' functional outcome. This quantitative meta-analysis of functional imaging studies of episodic encoding and retrieval tests the prediction that these deficits are most consistently associated with dysfunction in the prefrontal cortex.

Method: Activation likelihood estimation (ALE) was used to perform a quantitative meta-analysis of functional imaging studies that contrasted patients with schizophrenia and healthy volunteers during episodic encoding and retrieval. From a pool of 36 potential studies, 18 whole-brain studies in standard space that included a healthy comparison sample and low-level baseline contrast were selected.

Results: As predicted, patients showed less prefrontal activation than comparison subjects in the frontal pole, dorsolateral and ventrolateral prefrontal cortex during encoding, and the dorsolateral prefrontal cortex and ventrolateral prefrontal cortex during retrieval. The ventrolateral prefrontal cortex encoding deficits were not present in studies that provided patients with encoding strategies, but dorsolateral prefrontal cortex deficits remained and were not secondary to group performance differences. The only medial temporal lobe finding was relatively greater patient versus comparison subject activation in the parahippocampal gyrus during encoding and retrieval.

Conclusions: The finding of prominent prefrontal dysfunction suggests that cognitive control deficits strongly contribute to episodic memory impairment in schizophrenia. Memory rehabilitation approaches developed for patients with frontal lobe lesions and pharmacotherapy approaches designed to improve prefrontal cortex function may therefore hold special promise for remediating memory deficits in patients with schizophrenia.

FIGURE 1. Meta-Analytic Activation Map Showing Between-Group Differences for All Included Encoding Studies^a

^a Included studies are indicated in Table 1. Activation foci for contrasts of healthy comparison subjects > patients (red) and for regions where patients with schizophrenia > comparison subjects (green). Transaxial slices are spaced 4 mm apart and are presented according to neurological convention (right=right).

FIGURE 2. Meta-Analytic Activation Map Showing Regions in Which Activation in Healthy Comparison Subjects Is Greater Than in Patients With Schizophrenia, With or Without Encoding Strategies^a

^a The figure shows results for all selected encoding studies as illustrated in Figure 1 (red) and for a subset of incidental encoding studies in which patients were provided with encoding strategies (blue). Areas of overlap are indicated in purple, and the area of the dorsolateral pre-frontal cortex where groups continued to differ when patients were provided with encoding strategies is indicated by yellow arrows. Slices formatted as in Figure 1.

FIGURE 3. Meta-Analytic Activation Map Showing Between-Group Differences for All Selected Retrieval Studies^a

^a Included studies are indicated in Table 1. Activation foci for contrasts of healthy comparison subjects > patients (red) and for regions where activation in patients with schizophrenia > comparison subjects (green). Slices formatted as in Figure 1.