A case series of verbal semantic processing in primary progressive aphasia: Evidence from the N400 effect

Background – The semantic variant of primary progressive aphasia (PPA) is typically associated with a loss of semantic knowledge. Research on the semantic processing in the other clinical variants of PPA is, however, rather sparse and limited to off-line behavioural studies.Aims – This study aimed to investigate verbal semantic processing in patients with the three variants of PPA by the event-related potential technique. The presence, latency, amplitude, and/or topographic distribution of the N400 effect may be helpful in the diagnosis of PPA and its clinical variants and provide temporal information about semantic processing (disturbances) in the three variants of PPA. Methods & Procedures – The N400 effect was studied by a categorical word priming paradigm and a semantic anomaly paradigm at sentence level in eight persons with PPA(-plus) and 30 age-matched healthy controls. The mean amplitudes and onset latencies of the N400 effect were compared between each patient and the control group by two methods that are applicable in clinical practice namely visual inspection and Z-scores.Outcomes & Results – The N400 effect elicited by the categorical priming paradigm was only present in the two patients with the nonfluent variant of PPA. This effect was absent in the two patients with the semantic variant(-plus), two patients with the logopenic variant(-plus), one patient with the nonfluent variant-plus, and the patient with PPA not otherwise specified. The results of the N400 effect elicited by the semantic anomaly task at the sentence level were variable but differences in the presence, mean amplitudes, onset latencies, and/or topographic distributions of the effect were found in all patients with PPA(-plus) in comparison to the control group.Conclusions & Implications – The results of our study showed that the evaluation of the N400 effect might have an added value in the diagnostic process of PPA in general and in the differentiation of patients with the nonfluent variant from patients with the logopenic and semantic variants. Furthermore, our results indicate the presence of difficulties with retrieving stored semantic knowledge or semantic integration of a word in the preceding context in patients with the three variants of PPA. These findings might help the speech-language pathologist in determining individualized therapy goals and indicate that it might be helpful to focus on verbal semantic processing in language therapy in patients with the three variants of PPA and not only in patients with the semantic variant.