Causes of intraindividual variability in reaction times: a neurophysiologically oriented review and a new suggestion

Inter-individual variation in reaction times, and its negative correlation with intelligence, has long been established. Marked intra-individual variability in reaction times, within a given RT paradigm, has also been known for decades but, in contrast, is still unexplained and is usually ignored. Jensen [Jensen(1980)Bias in mental testing. New York: Free Press;Jensen, (1982)Reaction time and psychometric g. In H. J. Eysenck (Ed.), A model for intelligence. Berlin: Springer;Jensen, (1992)The importance of intraindividual variation in reaction time, Personality and Individual Differences, 13, 869-881.] has proposed an ‘‘oscillation model’’ in which a reaction stimulus-induced evoked potential interacts with ‘‘brain waves’’ (EEG) to make RTs vary from trial to trial within one test session. Anderson [Anderson, B. (1994) Speed of neuron conduction is not the basis of the IQ-RT correlation: Results from a simple neural model, Intelligence, 19, 317-324.] and Anderson and Donaldson [Anderson and Donaldson, (1995)The backpropagation algorithm: Implications for the biological bases of individual differences in intelligence, Intelligence, 21, 327-345.] have used simple computer models of connected neurons and neural nets, respectively, varying the values of certain parameters such as synaptic failure rates, to explain RT variability. I discuss these models and argue that they do not explain the observed variability. A new possible explanation is proposed here: intra-individual RT variability follows from variable (within a S, between individual RT trials) cortical pathway length (CPL), where CPL is the total length of the neural pathways in the cerebral cortex which an RT-induced evoked potential must follow in executing a correct RT response. This is an extension of the inter-individual CPL variability proposed by Reed and Jensen [Reed and Jensen, (1993) Choice reaction time and visual pathway nerve conduction velocity both correlate with intelligence but appear not to correlate with each other: Implications for information processing, Intelligence, 17, 191–203.].