This Article Abstract Full Text (PDF) Services Download to citation manager Citing Articles Citing Articles via HighWire PubMed PubMed Citation

Redundancy Gain and Coactivation in Bimodal Detection: Evidence for the Preservation of Coactive Processing in Older Adults

¹ Department of Psychology, The University of Akron, Ohio.
² NASA Ames Research Center, Moffett Field, California.

Address correspondence to Barbara Bucur, Box 2980, Duke University Medical Center, Durham, North Carolina 27710. E-mail: bb{at}geri.duke.edu

	Abstract

TOP Abstract Methods Results Discussion References

 
Previous investigations of adult age differences in the redundant signals effect suggest that both older and younger adults benefit from the presentation of redundant information. However, age deficits in divided attention may cause older adults to process redundant information in a different manner. In the present experiment, we tested between two competing explanations for the redundant signals effect: separate activation and coactivation. To investigate this issue, we used a bimodal detection task in which the auditory signal was a 1000-Hz tone and the visual signal was an asterisk. Both age groups showed significant violations of Miller's race model inequality, providing evidence for coactivation. These results suggest that, despite age-related deficits in divided attention, the ability to coactivate information from bimodal signals is spared with increased age.

APERVASIVE finding in the cognitive aging literature is that even relatively healthy older adults experience age-related cognitive decline across a variety of task conditions (Madden, 2001). Further, age-related decline in visual tasks tends to be most pronounced when some division of attention is required (Madden & Whiting, 2004). Although older adults exhibit age-related deficits in divided attention tasks, research has shown that older adults do benefit from the presentation of redundant information. In a reaction time (RT) task in which participants are instructed to press the corresponding key whenever the letter K or N is presented, participants are much faster responding to the presentation of two Ks compared with the presentation of a single K (Allen, Groth, Weber, & Madden, 1993; Allen, Madden, Groth, & Crozier, 1992; Allen, Weber, & Madden, 1994). This effect, the redundant signals effect (RSE; see, e.g., Miller, 1982), is significantly larger for older adults, indicating that this group shows a greater benefit from the presentation of redundant information compared with younger adults. The question that remains is this: Can the same model be used to explain the facilitation associated with redundant signals for both younger and older adults?

Researchers have proposed two models to explain the RSE: separate activation models and coactivation models. Separate activation models, also called race models, state that facilitation from redundant targets occurs because responding is based on whichever channel completes processing first. Race models assume that activation from multiple channels cannot combine to produce a single response; only information from the fastest channel receives further processing. Coactivation models, in contrast, assume that activation from multiple inputs, or channels, summate (coactivate) to produce a single response (Miller, 1982). The level of activation necessary to initiate a response builds more quickly from the presentation of multiple signals than from the presentation of a single signal. Therefore, coactive processing of redundant stimuli allows for faster responses than are possible in a race model with independent channels. To test for separate activation (race model) versus coactivation, Miller (1982) derived the race model inequality (Equation 1):

The test of the race model inequality is based on the cumulative distribution functions (CDFs), which show the probability, P, that a response is made by any given time. In Equation 1, t is the time since display onset, and S₁and S₂ represent signals on different modalities (e.g., auditory and visual). In a task in which the signals are a 1000 Hz tone and an asterisk, for example, the left side of the equation represents the CDF for trials with redundant signals (e.g., a tone and asterisk presented simultaneously). The two terms on the right represent the sum of the CDFs for the two types of single-signal trials. In essence, this equation says that, even under the most favorable conditions, a race model cannot produce responses to redundant-signal trials that are faster than the fastest response from single-signal trials. If this race model inequality is violated, then the race model can be rejected. Such a result would instead provide evidence for coactive processing of the two signals (e.g., Miller, 1982).

Although Allen and colleagues (1992, 1993, 1994) found a significantly larger RSE for older than younger adults, these researchers did not test for coactivation. (The tasks used by Allen and colleagues are not appropriate to test for coactivation. Mordkoff and Yantis, 1991, provided evidence that the coactivation occurring with redundant features within the same dimension results from biased contingencies, i.e., correlations between certain feature pairings, favoring faster responding to the redundant target trials. Allen et al. used one or two instances of the letters K and N; therefore, we would not expect violations of the inequality with this task.) It may be possible for older adults to show greater facilitation from separately activated channels (because their mean RTs are slower and have more room to benefit), yet not produce coactivation from redundant signals. Indeed, according to Miller (1982), the locus of coactivation is at the decision (central) stage of processing, a stage in which large age deficits on a variety of tasks have been observed (e.g., Cerella, 1985). To investigate these competing explanations for the RSE in older adults, we selected the bimodal detection task used by Miller. Consistent with Miller's work, we expect that a model of coactivation will provide an explanation for the RSE in younger adults. For older adults, however, a model of separate activation may provide the best explanation for the RSE because of divided attention deficits along with generalized slowing during central processing stages.

	METHODS

TOP Abstract Methods Results Discussion References

 
Participants
Twenty community-dwelling, healthy older adults between the ages of 60 and 82 (M = 67.7) and 20 younger adults between the ages of 18 and 34 (M = 23.0) from The University of Akron participated. Participants had corrected near visual acuity of at least 20/40. Older adults passed a standard auditory acuity screening (American Speech and Hearing Association, 2002), had none of the exclusionary conditions listed on the Christensen Health Screener (Christensen, Moye, Armson, & Kern, 1992), and scored at least 27 on the Mini-Mental State Exam (Folstein, Folstein, & McHugh, 1975).

Older adults scored significantly higher on the Mill Hill Vocabulary Test (Raven, Raven, & Court, 1997), t(38) = –2.96, p =.005, than did the younger adults, whereas younger adults scored significantly higher on the Wechsler Adult Intelligence Scale–Revised Digit Symbol Substitution Task subscale (Wechsler, 1981), t(38) = 6.52, p <.01. There were no differences between the age groups on the number of years of completed education or self-reported health.

Stimuli and Procedures
The visual stimulus was a white asterisk centered on a black background, which subtended 2° of visual angle. The auditory signal was a 1000-Hz tone presented at 70dB.

Each trial began with a 750-ms presentation of a central fixation cross. After a 250-ms interval, for trials containing signals, the signal was presented for 150 ms. For trials not containing signals, a blank screen was presented for 150 ms. There were 300 signal trials, which consisted of 100 trials of the asterisk, 100 trials of the tone, and 100 redundant trials containing both the asterisk and the tone. In addition, 300 no-signal trials were included. Within each block, stimuli were presented in a pseudorandom order. Trials on which errors were made were rerun at the end of each block. Participants completed a block of 24 practice trials followed by 10 blocks of 60 experimental trials.

Participants pressed the spacebar using the index finger of their dominant hand, as quickly as possible, whenever they saw an asterisk, heard a tone, or both. On all other trials (containing no signal), they were instructed not to respond.

	RESULTS

TOP Abstract Methods Results Discussion References

 
Participants in both age groups made very few errors (younger adults M =.024; older adults M =.021) and did not differ significantly. Prior to conducting the analyses, we removed all error trials.

Reaction Time Analyses
Using a split-plot analysis of variance, we found that neither the main effect of single-signal type (tone vs asterisk) nor the Age group x Single-signal type interaction were significant; therefore, we used the mean of the two single signals for the redundancy gain analysis.

The results of the redundancy gain analyses revealed a significant main effect of age group, F(1, 38) = 33.72, p <.001, and redundancy, F(1, 38) = 382.02, p <.001, as well as a significant Age group x Redundancy interaction, F(1, 38) = 7.97, p <.01, indicating a larger RSE for older adults. When using a proportional measure to control for the slower RTs of older adults in the single-signal condition (see Hartley, 1993 for details), we found that older adults, relative to younger adults, did not show a disproportionate improvement as a result of redundancy (older = 19%; younger = 18.5%). When testing for the RSE using the fixed favored dimension test (e.g., Mordkoff & Yantis, 1993), we found that the Age group x Redundancy interaction was not significant. (Several researchers, such as Biederman & Checkosky, 1970 and Mordkoff & Yantis, 1993, have suggested that the RSE could simply be an artifact of faster responding to a preferred single signal, either auditory or visual. On single-signal trials, the preferred signal is present on only half the trials. Both signals are present on redundant-signal trials, so responding to redundant-signal trials may simply be faster because the preferred signal is present on each trial. The fixed favored dimension test used the mean of only the fastest of the single-signal trials for each age group, resulting in a more conservative test of redundancy gain.)

Coactivation Analyses
Each participant performed 10 redundant-signal trials in each block. We first sorted these 10 RTs in ascending order of latency to estimate 10 percentiles (5th through the 95th at 10% intervals). We then averaged these numbers across blocks and participants to produce a composite CDF for the redundant-signal condition. To produce the sum of the two single-signal CDFs, we first pooled together the auditory-only trials and visual-only trials. Then we estimated the 10 percentile points based on the fastest 10 of the 20 trials within each block of each participant (see Miller, 1982). We averaged these 10 percentile points across blocks and participants to produce a composite CDF. Figure 1 shows that, for each age group, 8 of the 10 percentiles (5th through the 75th percentiles) showed a violation of the race model inequality (i.e., the redundant signal CDF was to the left of the sum of the single-signal CDFs). To determine if these violations were statistically significant, we conducted correlated groups t tests at each percentile. All eight violations were statistically significant for both age groups (ps <.05); therefore, we can reject separate activation models in favor of coactivation models.

View larger version (16K): [in this window] [in a new window]   Figure 1. Cumulative distribution functions (CDFs) for the sum of the single signals (solid curve) and for redundant signals (dashed curve). Miller's race model inequality is violated if the dashed curve is to the left of the solid curve at any percentile

	DISCUSSION

TOP Abstract Methods Results Discussion References

 
We conducted the current research to investigate possible age-related deficits in coactivation by using a bimodal detection task. If older adults experience deficits in divided attention, these impairments may in turn cause deficits in coactivating stimulus information. Therefore, although older adults show an RSE, we predicted that a separate activation model would provide a better explanation for this facilitation.

There were two main findings in this experiment. First, consistent with other research and our predictions, we found a significant Age group x Redundancy interaction, showing a larger RSE for older adults. However, when using a proportional measure, we found that older adults did not show a disproportionate improvement associated with redundancy. Although previous research has found a significant Age group x Redundancy interaction (e.g., Allen et al., 1992), in the current experiment we obtained such an effect only when using the mean of the two single-signal trials. When the fastest signal for both age groups (the tone) was used, the interaction was eliminated. This may have occurred from our attempts to make the auditory signal especially salient to compensate for age-related deficits in auditory acuity (Scialfa, 2002).

Inconsistent with our predictions, the second main finding is that older adults do exhibit violations of the race model inequality, thus providing evidence for coactivation. To investigate the role of generalized slowing on coactivation, we conducted a Brinley plot analysis using the mean RT at each of the 10 percentiles for the sum of the single-signal trials and the redundant-signal trials. Although this analysis revealed that the older adults' data could be characterized as a highly linear function (R² =.99) of the younger adults' data, the effects of generalized slowing appear to be reduced due to the presence of redundant signals.

Taken together, these results show that, because of generalized slowing, older adults do exhibit age-related deficits when they divide their attention between two different stimulus modalities; however, these were not sufficient to prevent the coactivation of bimodal stimuli. In fact, for both age groups, the amount of facilitation due to redundancy was greater than that predicted by a simple race model. Thus, although older adults do experience deficits in divided attention, they are able to process redundant information in qualitatively the same manner as younger adults. What is particularly interesting about this conclusion is that coactivation is assumed to occur at the decision stage—a stage thought to show large age deficits in many tasks (e.g., Cerella, 1985). Indeed, the current study represents the first empirical test that older adults show evidence for the preservation of a highly efficient form of parallel processing—coactivation.

	Acknowledgments

 
Barbara Bucur is now at Duke University Medical Center, The Center for the Study of Aging and Human Development. Support for her was provided by an American Psychological Association Science Directorate Dissertation Research Award, an Ohio Research Council on Aging Student Fellowship Award, and Grant T32 AG00029 from the National Institute on Aging.

We thank Ann McCarthy and Jeremy Grabbe for data collection and technical assistance. Thanks also go to David Madden, Kevin Kaut, Roberta De Pompei, Kathleen Shonk, and Ethel Bucur for helpful comments. This article is dedicated to the memory of Michael Bucur.

	Footnotes

 
Decision Editor: Thomas M. Hess, PhD

Received for publication July 13, 2004. Accepted for publication May 2, 2005.

	References

TOP Abstract Methods Results Discussion References

 

• Allen, P. A., Groth, K. E., Weber, T. A., Madden, D. J. (1993). Influence of response selection and noise similarity on age differences in the redundancy gain. Journal of Gerontology: Psychological Sciences, 48,P189-P198.
• Allen, P. A., Madden, D. J., Groth, K. E., Crozier, L. C. (1992). Impact of age, redundancy, and perceptual noise on visual search. Journal of Gerontology: Psychological Sciences, 47,P69-P74.
• Allen, P. A., Weber, T. A., Madden, D. J. (1994). Adult age differences in attention: Filtering or selection? Journal of Gerontology: Psychological Sciences, 49,P213-P222.
• American Speech and Hearing Association. (2002). Preferred practice patterns for the profession of speech language pathology: Audiologic screening performed by speech language pathologists. Washington, DC: Author.
• Biederman, I., Checkosky, S. F. (1970). Processing redundant information. Journal of Experimental Psychology, 83,486-490.
• Brinley, J. F. (1965). Cognitive sets, speed and accuracy of performance in the elderly. In A. T. Welford & J. E. Birren (Eds.), Behavior, aging, and the nervous system (pp. 114–149). Springfield, IL: Charles C. Thomas.
• Cerella, J. (1985). Information processing rates in the elderly. Psychological Bulletin, 98,67-83.[Medline]
• Christensen, K. J., Moye, J., Armson, R. R., Kern, T. M. (1992). Health screening and random recruitment for cognitive aging research. Psychology and Aging, 7,204-208.[Medline]
• Folstein, M. F., Folstein, S. E., McHugh, P. R. (1975). Mini-Mental State: A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research, 12,189-198.[Medline]
• Hartley, A. A. (1993). Evidence for the selective preservation of spatial selective attention in old age. Psychology and Aging, 8,371-379.[Medline]
• Madden, D. J. (2001). Speed and timing of behavioral processes. In J. E. Birren & K. W. Schaie (Eds.), Handbook of the psychology of aging (5th ed., pp. 288–312). San Diego: Academic Press.
• Madden, D. J., Whiting, W. L. (2004). Age-related changes in visual attention. In P. T. Costa & I. C. Siegler (Eds.), Recent advances in psychology and aging (pp. 41–88). Amsterdam: Elsevier.
• Miller, J. (1982). Divided attention: Evidence for coactivation with redundant signals. Cognitive Psychology, 14,247-279.[Medline]
• Mordkoff, J. T., Yantis, S. (1991). An interactive race model of divided attention. Journal of Experimental Psychology: Human Perception and Performance, 17,520-538.
• Mordkoff, J. T., Yantis, S. (1993). Dividing attention between color and shape: Evidence of coactivation. Perception and Psychophysics, 53,357-366.[Medline]
• Myerson, J., Ferraro, F. R., Hale, S., Lima, S. D. (1992). General slowing in semantic priming and word recognition. Psychology and Aging, 7,257-270.[Medline]
• Raven, J., Raven, J. C., Court, J. H. (1997). Mill Hill Vocabulary Scale (1998 ed.). Oxford, England: Psychologists Press.
• Scialfa, C. T. (2002). The role of sensory factors in cognitive aging research. Canadian Journal of Experimental Psychology, 56,153-163.
• Wechsler, D. (1981). Wechsler Adult Intelligence Scale–Revised. New York: Psychological Corporation.

This article has been cited by other articles:

				P. Lemaire and M. Lecacheur Aging and Numerosity Estimation J. Gerontol. B. Psychol. Sci. Soc. Sci., November 1, 2007; 62(6): P305 - P312. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Services Download to citation manager Citing Articles Citing Articles via HighWire PubMed PubMed Citation