Study of the relationships between evoked potentials, inspection time and intelligence
This thesis examines the relationships between average evoked potentials (AEPs), inspection time (IT) and intelligence, by relating visual AEPs evoked by the stimuli of an inspection time task (and by similar stimuli for which an IT-type discrimination was not required) to subjects' IT estimates, and to their intellectual ability as measured by intelligence tests.Earlier work on IT and AEP in relation to intelligence is reviewed.Evoked potentials were collected while subjects were performing an IT (or an equivalent) task. In Experiments 1 & 2, it was found that a measure of the P200 ("P200r"· defined in Chapters 3 and 4) of AEPs to the IT task stimuli correlated significantly with IT (Pearson: r=0.57, p<.05). Experiment 3 replicated this finding (r=.44, p<.05). Also, Experiment 3 found that P200 latency could be related to IT (r=.55, p<.05). These results were obtained again in Experiment 6 (r=.645, p<.0005. for P200T; r=.442, p<.005 ,for P200 latency). It is inferred from these results and those reported by other authors that P200 reflects the process of encoding or transferring information from a sensory register into short-term memory (STM). Further, it is argued that inspection time indexes the rate at which sensory input is sampled in the initial stages of information processing.Several techniques were used to examine the relationship between the P300 component and IT. P300 latency was not closely related to IT, but P300 amplitude corre lated positively and significantly with IT (Experiment 6). P300 amplitude reflected subjects' confidence in their performance of a task, and this result suggests that subjects' choice of a more stringent criterion of confidence for their judgements may contribute to their longer measured ITs.Another factor which may also play a part in subjects' performance in the IT task is the process of anticipation of task stimulus. When the warning period was extended from 500 msec to 1800 msec (in Experiments 4 and 5), it was found that the strength of anticipation, as Indexed by the amplitude of contingent negative variation (CNV), correlated positively with IT. This suggests that strong anticipation may, under these conditions, handicap subjects' performance on the IT task.Experiment 6 examined the relationships between 10. IT and the measures of AEPs previously found to correlate with IT. Each subject was presented with his/her IT-duration stimulus. Half of the presentations were designated as "task-loaded" trials requiring an IT response. and the other half the "task-free" trials requiring no IT response, with the two kinds of trials randomly intermixed. In each trial, subjects were asked to give a reaction-time response to a visual signal following the IT-duration stimulus.In this experiment as expected, IT correlated negatively with intelligence test scores. The previously identified parameters of AEPs to the IT-duration stimulus with task requirements correlated with IT, but not 10; these therefore reflect task-specific individual differences. In contrast measures of the P200 to the digit stimuli which identified the nature of a trial (i .e. with or without IT-task requirements) did correlate with 10. and reflect individual differences related to general cognitive ability. Subjects' inspection time also correlated with the non-specific AEP differences.In the light of the results described above, the IT-10 relation may be seen to depend on a general speed factor. reflecting the process of encoding sensory input into STM from a sensory register. The higher the speed of this encoding process (i.e. smaller values of the P200 temporal measures), the shorter the inspection time and the higher the intelligence test score.