Cross-modal negative priming and interference in selective attention

Bulletin of the Psychonomic Society 1993, 31 (I) , 45-48 Cross-modal negative priming and interference in selective attention JON DRIVER University of Cambridge, Cambridge, England and GORDON C. BAYLIS University of California, San Diego, California This study examined whether negative priming and Stroop interference can be observed crossmodally, resulting from auditory distractors that are related to subsequent visual targets. Negative priming was found in subjects who remained unaware of the contingency between distractors and subsequent targets, whereas subjects who reported the contingency showed facilitatory priming. Stroop interference from incongruent auditory distractors was observed across all subjects and was replicated in a second experiment which also showed that interference declined when a particular distractor was repeated. These data suggest that an inhibitory mechanism of selection may operate to prevent response to auditory distractors, similar to the inhibitory selection mechanisms proposed for vision. They also demonstrate the reality of cross-modal Stroop effects, contrary to the findings of Miles, Madden, and Jones (1989). Studies of human selective attention are concerned with how people select a subset of available perceptual information to control their responses . Although the level of processing at which selection takes place remains controversial , most authors agree that ignoring a stimulus is a passive process . Neisser (1976 , p. 87) provides a strong statement of this view: " What then happens to unattended information? In general nothing happens to it. . . . We simply don 't pick it up. " An alternative view holds that unattended information is actively inhibited (Neill, 1977; Tipper & Driver, 1988). The negative priming effect provides prima facie support for such inhibition. This effect was first observed by Neill (1977) in a variant of the Stroop task. When the color to be named on trial n + 1 is the same as the color word that has been ignored on trial n, a delay is observed; this delay is taken to reflect continued inhibition of some representation of the previously ignored word. Analogous effects have since been observed for line drawings or digits, with selection being cued by color (Driver & Tipper , 1989; Tipper, 1985; Tipper & Driver, 1988). Tipper and Driver (1988) found negative priming when an ignored drawing was followed by an attended visual word that was the name of the ignored object. The lack of any physical resemblance between the drawing and the word suggests that this effect arose at an abstract level of representation. The present study examines whether negative priming can be observed between an auditory distractor and a subsequent visual target. J.D .' s research is supported by the Medical Research Council (U.K.), G.C .B. was funded by U.S. Office of Naval Research Contract NOOOI488-K-Q281. Correspondence concerning this art icle may be addressed to J. Driver, Department of Experimental Psychology . University of Cambridge. Downing Street . Cambridge CB2 3EB, England . 45 In addition to examining the impact of auditory distractors on response to a subsequent visual target , in our first experiment we investigated whether auditory distractors can interfere with responses to a simultaneous visual target. The existence of Stroop (1935) interference from one modality to another has a controversial history . Broadbent (1982, p. 284) asserted that " it is hard to imagine that selectivity cannot be reduced to one sense excluding the other. " More recently , Miles , Madden, and Jones (1989) argued on theoretical grounds that cross-modal Stroop effects should never be observed. Contrary to these suggestions, there is some positive evidence for the existence of cross-modal interference effects . For example, Greenwald (1972) had subjects name a series of visual digits while they ignored a concurrent stream of auditory distractors . He reported a Stroop effect from auditory distractors that were incongruent digits. However, the interfering digit was always preceded throughout the series by auditory clicks . An orienting response to the first speech sound in the series might therefore have produced the interference, as noted by Broadbent (1982 , p. 284). To preclude this possibility , we compared the interference from sequences of auditory digits on visual digit naming with the distraction produced by sequences of speech that did not include digits. Greenwald (1972) also observed an effect that it is tempting to interpret as cross-modal negative priming. When an auditory digit was repeated throughout a series , response to the visual target at the end of the series was delayed if it corresponded to the auditory digit. Unfortunately, the successive repetition of the digit as a distractor throughout the series means that habituation to the auditory distractors may occur . Thus the increased response latency might reflect cross-modal habituation, rather than the inhibition that is held to produce negative priming (see Lorsch , Anderson, & Well , 1984; Reisberg , Baron , & Copyright 1993 Psychonomic Society, Inc. 46 DRIVER AND BAYLIS Kemler, 1980). In our first experiment, we extended the Greenwald (1972) paradigm to include the crucial negative priming condition , in which an unrepeated auditory distraetor corresponds to the visual digit at the next position. EXPERIMENT I Method Subjects. The 24 subjects (17 female, 7 male) were paid volunteers from the Oxford subject panel. All reported normal or corrected-tonormal acuity and normal hearing . Apparatus and Materials. The stimuli were generated on an Acorn BBC microcomputer (Model B +). Auditory distractor stimuli were generated with an Acorn BBC speech system, which consists of a speech processor and a ROM with prestored words and word parts, all in the same voice. The stimuli employed were the digits 1,2, 3, 4, 5, 6, 8, and 9, as well as a nonsense vowel sound ("uh"). They were presented to both ears at about 70 dB over a pair of monaural headphones connected to a Marantz Superscope CD-330 cassette recorder, which served as an amplifier. Visual target stimuli were the digits I, 2, 3, 4, 5, 6, 8, and 9 presented in Teletextdouble size characters on a MicrovitecCub 895 Monitor. At a viewing distance of 90 em, the average visual angles were .4 0 horizontal and .7 0 vertical. Additional visual materials were a fixation cross and a hash mark used as a visual mask. The stimulus series consisted of nine successive visual targets and nine auditory distractors, which were each presented concurrently with a visual target. Naming reaction times (RTs) to the visual stimuli in Positions 2-9 for each series were taken in milliseconds with the use of a TTL voice key. A hand-held microswitch was used by subjects for starting each stimulus series . Design . A within-subjectsdesign was employed, in which the independent variable was the relationship of the auditory distractors in (...truncated)