SYLVAN, ANITA.

Previous research has shown that priming of digits is a
function of the ordinal distance of the target digit from
the priming sequence and that under certain conditions this
effect may be automatic. The present experiment, which
employs a choice RT task, was designed to test automaticity
by manipulating subjects' expectancies and the time
available for processing the priming sequence.
Schematically valid primes (e.g., "1,2,3" preceding "4")
were presented on 30%, 50%, or 70% of the trials to a given
subject. Invalid primes (e.g., "5, 6, 7" preceding "4")
were presented on the remaining trials. There was a highly
significant effect of priming modified by interactions with
exposure time and frequency. Employing a cost-benefit
analysis to the frequency factor, it was found that the most
frequently occurring prime resulted in decreased response
latency but the less frequently occurring prime did not
result in an increase in response latency (benefit wihout
cost). On this basis, it is concluded that frequency
influences an automatic process in this experiment. Since
frequency interacts with priming, it is therefore argued
that priming also affects an automatic process.