Miotto, E.C., Savage, C.R., Evans, J.J., Wilson, B.A., Martin, M.G., Balardin, J.B., . . . & Amaro Junior, E. (2013). Semantic strategy training increases memory performance and brain activity in patients with prefrontal cortex lesions. Clinical Neurology and Neurosurgery, 115, 309-316.

To investigate the neural correlates of semantic strategic training in patients with distinct prefrontal cortex lesions

In the 30-minute semantic organizational strategy training, subjects were taught three steps to improve memory: (1) organize words into categories, (2) memorize words, and (3) retrieve words by category. Subjects were required to undergo a minimum of five training trials and to demonstrate that they had learned how to apply the semantic organizational strategy. Pre- and post-test evaluations consisted of subjects undergoing functional magnetic resonance imaging (fMRI) scanning immediately prior to and after the semantic strategy training.

• N = 21      
• MEAN AGE = 45.7 years for bilateral orbitofrontal cortex (BOFC) patients and 39.57 years for dorsolateral prefrontal cortex (LPFC) patients
• MALES: 57%, FEMALES: 43%
• KEY DISEASE CHARACTERISTICS: 21 patients with confirmed primary brain tumor (PBT) located in the prefrontal cortex, specifically the LPFC or BOFC
• OTHER KEY SAMPLE CHARACTERISTICS: At least six months post-chemotherapy and radiation therapy; time since surgical resection ranged from 6 months to 10 years (mean = 2.65 years; SD = 2.4 years); all subjects were non-aphasic, right-handed, and free of other neurological or psychiatric conditions.

• SITE: Not stated/unknown  
• SETTING TYPE: Not specified    
• LOCATION: Not stated but ethics approval was received from the University of Sao Paulo Medical School.

Exploratory study using a pre- and post-test design in a convenience sample

• 3 word lists of 16 words each: unrelated and related non-structures, and related structure word lists for cortical activation during scanning
• fMRI scan

Both subject groups recalled more words (p < 0.001) and organized more words together (p < 0.001) after semantic training. For the BOFC group, there were significant differences (p < 0.01) after the intervention of increased activation: (1) right cuneus and posterior cingulate gyrus, and left precuneus during the related structure word list trial and (2) left precentral gyrus, postcentral gyrus, inferior frontal gyrus, and insula during the unrelated word list trial. For the LPFC group, there were significant differences (p < 0.01) after the intervention, in multiple cortical areas: (1) increased activation in the left inferior frontal gyrus, precentral gyrus, and insula during the related non-structured word list trial (2) increased activation in the left inferior frontal gyrus, precentral gyrus, and anterior cingulate during the related structure word list trial, and (3) decreased activation in the left supramarginal gyrus, precuneus, superior temporal gyrus, and superior parietal lobe during the related structure word list.

The intervention increased use of categorizing words, thereby improving memory performance in subjects with prefrontal PBT. Changes in cortical activation suggest intervention application and recruitment of compensatory brain mechanisms for memory tasks. Further study with longitudinal measures is warranted to demonstrate continued cortical activation and intervention application during memory tasks.

• Small sample (< 30)
• Risk of bias (no control group)
• Findings not generalizable
• Intervention expensive, impractical, or training needs
• Other limitations/explanation: fMRI scanning for pre- and post-measures is expensive, is impractical, and requires specialized training and equipment. This study was performed in a homogenous sample.

This study demonstrated the application of a quick single session semantic training intervention aimed at improving verbal memory performance in those with known neuropathology.