Intersensory Redundancy: Infants’ Detection of Unimodal Properties of Events*

 

Lorraine E. Bahrick1, Robert Lickliter2, and Ross Flom1

 

1 Florida International University             2 Virginia Polytechnic Institute and State University

                           Department of Psychology                                             Department of Psychology

                                Miami, FL 33199                                                         Blacksburg, VA 24061

 

 

Abstract

Bahrick and Lickliter (2000) proposed an intersensory redundancy hypothesis which states that in early development 1) information presented redundantly across two sensory modalities recruits attention to redundantly specified properties of events at the expense of other properties, whereas 2) the same event presented to one sense modality alone selectively recruits attention to modality-specific aspects of the event and facilitates perceptual learning of those properties at the expense of others.   Prior research has supported the first tenet of this

hypothesis, but to date no test of the second tenet has been conducted. The present study thus tested the second tenet by assessing infants' detection of orientation, a unimodal visual property, under both unimodal and bimodal presentation conditions.  Five-month-old infants (N= 32) were habituated to films of a hammer tapping out a distinctive rhythm in one of two orientations (up versus down) under either a unimodal (visual) or a bimodal (auditory and visual) condition.   Results confirmed our hypothesis and indicated that following unimodal habituation infants showed significant visual recovery to the change in orientation (p<.02), but following bimodal habituation they showed no visual recovery. Together with prior results, these findings provide support for both tenets of the intersensory redundancy hypothesis and have important implications for the development of perception and cognition.

 

 Introduction

  Research indicates that infants perceive coherent, unified multimodal objects and events through different sense modalities even in the first months of life.  Little is known, however, about how they achieve such impressive intersensory capabilities at such young ages. Bahrick & Lickliter (2000) recently proposed an "intersensory redundancy hypothesis" to account for an important aspect of this developmental process.  The intersensory redundancy hypothesis posits that information experienced redundantly, and in temporal synchrony, across two sensory modalities selectively recruits infant attention to bimodally specified properties (such as tempo or rhythm) at the expense of modality-specific properties of events, whereas information experienced in one sense modality alone selectively recruits attention to modality-specific properties of events (such as orientation, color, pitch, or timbre) at the expense of bimodally-specified properties.  Prior research has supported the first tenet of the hypothesis (Bahrick & Lickliter, 2000; Bahrick, Lickliter, & Flom, 2000).  These studies demonstrated that infants who were habituated to a hammer tapping out a rhythm at a particular tempo showed visual recovery to a new rhythm at 5 months of age, and to a new tempo at 3 months of age, when the events were experienced in two sense modalities, but not when they were experienced in one modality alone.  To date, however, no test of the second tenet of the redundancy hypothesis had been conducted.  The present study thus assessed infants' detection of orientation, a unimodal visual property, under both unimodal and bimodal presentation conditions. We predicted that infants would discriminate changes in orientation under unimodal but not bimodal conditions.

 

 

Methods

Thirty-two 5-month-old infants were habituated, in an infant-controlled procedure, to films of a hammer tapping out a rhythm (at the rate of 110 beats per minute) used in the prior studies. The movements of the hammer depicted one of two orientations (upward versus downward).  Either the hammer hit downward against a wooden floor, or it hit upward against a wooden ceiling.  These events are depicted in Figure 1. The event could be conveyed unimodally by watching the hammer, or it could be conveyed bimodally by watching the hammer and hearing its impact sounds. Infants were randomly assigned to the unimodal (visual) or bimodal (audio-visual) habituation conditions and half the infants in each condition received the hammer striking upward and half received the hammer striking downward. Following habituation and two no-change post habituation trials, infants received two test trials depicting a change in orientation under their respective conditions.  Visual recovery (the difference between the mean number of seconds looking during the two test trials versus the two no-change post habituation trials) was assessed and served as the measure of discrimination. Interobserver reliability for visual recovery was .97.

 

 

Figure 1: Static images of the dynamic events depicting the two different orientations.

 


 

      

                              

 

Results

The dependent variable, visual recovery, was calculated by subtracting the mean number of seconds looking during the two no-change post habituation trials from the mean number of seconds looking during the two test trials. The results are depicted in Figure 2.  Infants showed significant visual recovery to the change in orientation under conditions of unimodal visual habituation and testing, t(15) = 2.64, p = .018.  However, under conditions of bimodal audio-visual habituation and testing, infants showed no evidence of detecting a change in orientation, t(15) = .87, p >.1. Thus, infants discriminated changes in orientation, a modality-specific property, following unimodal visual exposure, but not following bimodal audio-visual exposure.

Conclusions

 These findings demonstrate that 5-month-old infants are able to perceive a change in the modality-specific property of orientation when events are experienced unimodally, whereas when events are experienced bimodally, they show no evidence of sensitivity to orientation. Apparently, the addition of the soundtrack created intersensory redundancy and selectively recruited attention to redundantly specified properties (such as rhythm and tempo; see Bahrick & Lickliter, 2000; Bahrick, Lickliter, & Flom, 2000), at the expense of unimodal properties such as orientation.   Taken together with our prior research, these findings support both tenets of the intersensory redundancy hypothesis.  That is, in early development 1) bimodal audio-visual exposure to an event selectively recruits attention to redundantly specified properties of the event at the expense of modality-specific properties, and 2) unimodal auditory or visual exposure to an event selectively recruits attention to modality-specific properties of the event at the expense of amodal properties.  These findings have important implications for the development of perception and cognition and suggest that in a world of predominately multimodal stimulation, bimodally specified properties are selectively attended, fostering perception, learning and memory of those aspects of stimulation prior to other aspects.  These conclusions are also consistent with those from research with animal infants (Honeycut & Lickliter, in press; Sleigh, Columbus, & Lickliter, 1998) demonstrating a functional distinction between unimodal and bimodal sensory stimulation during the prenatal and early postnatal periods.

 

References

     Bahrick, L.E. & Lickliter, R. (2000). Intersensory redundancy guides attentional selectivity and perceptual learning in infancy.  Developmental Psychology, 36, 190-201.

      Bahrick, L.E, Lickliter, R., & Flom,  R. (2000). Intersensory redundancy guides perceptual learning: Discrimination of tempo in 3-month-olds. Paper presented at the 2nd annual Multisensory Research Conference, Tarrytown, NY.

      Honeycut, H. & Lickliter, R. (in press). Order-dependent timing of unimodal and multimodal stimulation affects prenatal auditory learning in bobwhite quail embryos. Developmental Psychobiology.

     Sleigh, M.J., Columbus, R., & Lickliter, R. (1998). Intersensory experience and early perceptual development: Postnatal experience with multimodal maternal cues affects intersensory responsiveness in bobwhite quail chicks. Developmental Psychology, 34, 215-223.

 


* Poster presented at the 33rd Annual Meeting of the International Society for Developmental Psychobiology,

New Orleans, LA November 1-4, 2000. This research was supported by a NIH grant, RO1 HD25669, awarded to the first author and a NIMH grant, K02 MH01210, awarded to the second author.

Address correspondences to Lorraine Bahrick: bahrick@fiu.edu