Date of Award


Degree Name

Doctor of Philosophy


Speech-Language Pathology


Anthony P. Salvatore


In this study, six individuals with aphasia and twelve participants with no brain damage responded to spoken sentence length messages using a modified version of the Revised Token Test (McNeil & Prescott, 1978) while cortical activation was recorded using event related potentials (ERP). ERP is a non-invasive imaging procedure that measures cortical activation reflected in the electrical activity that is produced at the level of the cortex in response to internal or external stimulus.

Participants were presented with a visual display that provided eight different token arrangements per trial that were displayed on the computer screen. Spoken messages that increased in length and grammatical complexity were presented via speakers. Participants were instructed to listen to the spoken message and respond by touching the visual choice that matched the spoken message on a touch screen for a total of 90 trials.

The task of semantic processing may be reflected in the latency and peak amplitude of the N400 ERP component. This component has been reported to reflect the processing of single spoken words in participants with no brain damage and individuals with aphasia (Friederici et al., 1999; Handy, 2005; Luck, 2005; Rugg & Coles, 1997). The N400 ERP component has not been well investigated in the processing of spoken sentence length messages in individuals with aphasia.

In order to investigate the mechanisms involved in the auditory comprehension of spoken sentence length messages, we need to identify where in that temporal process the individual is processing the message. Since ERP can be time locked to a specific event, it seems reasonable to use ERP to investigate the temporal characteristics of auditory comprehension of sentence length messages.

The present study compared the performance of individuals with aphasia and participants with no brain damage responding to spoken sentence length messages. The results show that the two groups performed differently in regards to correct response rate. The individuals with aphasia made more error responses than did the participants with no brain damage. The participants with no brain damage showed reduced accuracy as the complexity of the sentence increased while the individuals with aphasia maintained the same performance throughout the test. No statistically significant difference was found between the individuals with aphasia and the participants with no brain damage for the mean latency and amplitude of the P300 and N400 components. However, upon visual examination of the ERP waveforms, the individuals with aphasia exhibited longer ERP latencies of the N400 and the P300 than the participants with no brain damage. In addition, the individuals with aphasia displayed slightly increased amplitudes for the P300 ERP component. Furthermore, the cortical activation patterns between the two groups were different.The non-brain damaged individuals displayed some degree of activation in the left frontal electrodes for the N400 component while the individuals with aphasia exhibited activation in the frontal and right temporal electrodes with some degree of activation exhibited in the left temporal electrodes. Furthermore, the individuals with aphasia exhibit high areas of activation throughout the cortex for the P300 suggesting that the individuals with aphasia were recruiting all cortical areas available to them to compensate for damage to localized areas. The participants with no brain damage displayed activation in the frontal, right and left temporal electrodes. Cortical activation was also exhibited in the parietal electrodes consistent with the literature that suggests that the P300 is elicited primarily in the parietal electrodes.

The nature of activation also differed across the duration of each sentence in each of the two groups of participants. The individuals with aphasia showed reduced attentional activity at the start of each trial and an increased latency and reduced amplitude when processing the spoken message. The participants with no brain damage showed quicker attentional response to the onset of each trial and a shorter latency and increased amplitude when processing the sentence.

The implication of this study is that the individuals with aphasia processed spoken sentence length messages different from participants with no brain damage. Their attentional response was delayed and they required added processing time, but obtained a moderate degree of correct response rate.




Received from ProQuest

File Size

140 pages

File Format


Rights Holder

Patricia Lara