Interaction of encoding and retrieval in relative age judgments: An extension of the "crossover" effect.

Abstract

R. J. Audley and C. P. Wallis have demonstrated the existence of what they call the "crossover" effect in the speed of relative judgments. They showed that 5s are faster at choosing the brighter than the darker of two stimuli when both are relatively bright, but for two dark stimuli the choice of the darker is quicker. The present study extends this finding for relative age judgments in which 5s must choose the younger or older of two pictured adults or children. A model to account for this effect is proposed which assumes that 5s encode the instruction at a first stage of processing, encode the stimuli at an independent second stage, and compare these encoded representations at an independent third stage. Several related studies are discussed which provide support for these assumptions.

Comparative judgments with numerical reference points

“…The latter include judgments of relative pitch and of brightness ; e.g., people can choose the higher of two "high" tones more quickly than they can choose the lower, while they can choose the lower of two "low" tones more quickly than they can choose the higher. In addition, similar effects have been found with more complex comparisons such as judgments of the relative age of pairs of children vs adults (Ellis, 1972) and of the relative skin color (lighter or darker) of pairs of whites or blacks (Friend, 1973). In some experiments the congruity effect takes the form of a "crossover" interaction, so that the relative difficulty of the two questions is actually reversed at opposite ends of the scale.…”

Perceptual encoding in comparative judgments of race

“…This study investigated the perceptual encoding of racial characteristics by examination of data on comparative judgments along the black-white racial dimension. A model recently proposed by Ellis (1972) is used in the interpretation of the data. This model assumes that the time between stimulus presentation and response execution is occupied by a series of processing stages which contribute additively to the total reaction time (cf, Sternberg.…”

“…The model proposed by Ellis (1972) will predict the crossover effect if the result of Stage 1 is an encoding consistently in terms of the overall level of the stimulus pair. For example, in Experiment 4 of Audley and Wallace (1964) (which actually produced a crossover), the background illumination was intermediate to the levels of the stimulus pairs-this permitted the darker stimuli and the lighter stimuli to be consistently encoded in the appropriate linguistic terms.…”

“…For example, in Experiment 4 of Audley and Wallace (1964) (which actually produced a crossover), the background illumination was intermediate to the levels of the stimulus pairs-this permitted the darker stimuli and the lighter stimuli to be consistently encoded in the appropriate linguistic terms. Similarly, in the relative age judgments (Ellis, 1972), the Ss' ages (college sophomores) were intermediate to the ages of the younger and older pairs of faces to be compared permitting consistent encoding. The funnel effect is obtained from the model when the result of Stage 1 is an inconsistent encoding of one of the stimulus pairs.…”

Expanding SNARC for temperature, negative numbers and intelligence

“…The semantic congruity effect has since been found in a variety of situations including pictorial comparisons of age (Ellis, 1972), skin colour (Friend (1973), audible comparisons of loudness (Banks and Root, 1979), numbers (Banks, Fujii, & Kayra-Stuart, 1976), and words representing length, quality, and temperature (Holyoak and Walker, 1976). The semantic congruity effect has also been previously found using the instructions "Smaller" and "Larger" for animal names, (Jamieson andPetrusic, 1975, Banks andFlora, 1977) as well as animal pictures (Banks and Flora, 1977).…”