As time-travelers turning the clock back about a half century, we might be surprised to discover that the importance of the link between auditory and cognitive processing had already been anticipated in the post-WWII era of hearing research and in the early days of audiology education and practice.

A diagram in one of the earliest audiology textbooks, Hearing and Deafness, showed the physical, anatomical, physiological, and psychological aspects of speech communication, including connections between information processing by the peripheral and central auditory systems and information processing at higher cognitive levels, including language, memory, and attention (Davis, 1964, in Davis & Silverman, 1970). Even earlier, seminal research on auditory attention by Collin Cherry (1953) and his famous writings about the "cocktail party" effect foreshadowed the current interests of hearing-aid manufacturers for whom a rebirth of interest in cognition has been inspiring the design of new hearing aids (Edwards, 2007).

The relationship between auditory and cognitive processing has been recognized as important for decades, but new technologies now enable us to understand it in a new way. Advances in cognitive neuroscience now provide answers to questions about those relationships, which earlier researchers did not have the tools to tackle. At the same time, advances in digital signal processing offer us rehabilitative options that were not possible with older technologies. These advances now compel audiologists to shift from a narrower focus on the ear and hearing to a broader consideration of the brain and listening (Pichora-Fuller, 2003; Pichora-Fuller & Singh, 2006).

The Audition-Cognition Link

Four main factors are motivating audiologists to explore the connection between audition and cognition (for discussions see Pichora-Fuller, 2006, 2007):

• The everyday challenges encountered by people living with hearing loss in the complex acoustic ecologies of the real world cannot be understood only in terms of hearing impairment (Kiessling et al., 2003)
• In multi-tasking situations, difficulty hearing may exacerbate or masquerade as reductions in cognitive performance, including problems with remembering and/or comprehending spoken language (Pichora-Fuller, Schneider, & Daneman, 1995; Schneider, Daneman, Murphy, & Kwong See, 2000)
• The ability of listeners, especially older adults, to use preserved cognitive abilities and supportive context to compensate for declines in the rapid processing of reduced sensory information offers hope for new rehabilitative interventions (e.g., Kricos, 2006; Pichora-Fuller, in press)
• Cognitive factors have recently been recognized to be important predictors of success with hearing aids, especially technologies with fast-acting, complex signal-processing (e.g., Edwards, 2008; Lunner, & Sundewall-Thorén, 2007)

An additional reason to link auditory and cognitive processing as we shape new best practices is the mounting evidence that auditory processing disorders may predict and accelerate the onset of symptoms of cognitive disorders such as dementia (e.g., Gates, Beiser, Rees, Agostino, & Wolf, 2002; Golding, Taylor, Cupples, & Mitchell, 2006). The first four factors indicate why the link between cognition and audition should be important to audiologists working with adults of any age; the additional reason is especially important to audiologists working with older adults.

Research Evidence

In daily life listeners constantly take in new "bottom-up" information using their senses, and they combine these inputs with "top-down" knowledge they have learned and stored in various brain regions (Craik, 2007). Listeners use the auditory system to accomplish many everyday functions including hearing, listening, comprehending, and communicating (see Kiessling et al., 2003); however, these functions also require more general cognitive operations such as attention, memory, and language. Simply put, our ears enable us to hear passively, but our brains enable us to use actively what we have heard for specific purposes.

As Davis and Silverman (1970) anticipated, despite the obvious importance of discovering how brain systems coordinate information when a person is engaged in complex behaviors such as communication, considerable research would be needed to develop the methods for studying the relationship between brain hardware and cognitive software. Fortunately, audiologists are now poised to apply exciting new knowledge about the interactions between auditory and cognitive processing that has been discovered using behavioral, physiological, and epidemiologic research approaches.

Behavioral Research

Because it is well-known that the prevalence of both auditory and cognitive deficits increases with age, it is not surprising that the behavioral research linking audition and cognition arose from research on adult aging. Both types of declines provide possible explanations for older adults' frequent reports of difficulty understanding speech in noise (CHABA, 1988). Initially, hearing researchers tried to distinguish whether the speech understanding problems of older adults were due primarily to auditory (peripheral or central) or cognitive deficits (Humes, 2008). However, recognizing a strong correlation between sensory and cognitive aging, leading cognitive researchers by the mid-1990s called for new approaches to investigate various hypotheses regarding the reasons underlying the strong correlations between auditory and cognitive aging (Baltes & Lindenberger, 1997; Lindenberger & Baltes, 1994).

One of the hypotheses was information degradation, which suggests that reduced quality of the sensory input will result in less efficient cognitive functioning. We have demonstrated that memory and comprehension suffer when the quality of the sound input is reduced by masking or by temporal distortion (Pichora-Fuller, 2007). In almost all studies this pattern of results is equivalent for younger and older adults (Schneider, Pichora-Fuller, & Daneman, in press). It seems that, regardless of age, a listener is more able to use the information that has been heard if the quality of the input is better. Thus, cognitive performance is optimal when listening is effortless and is reduced when listening is effortful.

In challenging conditions requiring effortful listening, such as the presence of multiple competing talkers, cognitive resources are presumed to be diverted to listening. As listening effort is increased, fewer cognitive resources remain available for remembering and/or comprehending what was heard. In such conditions, listeners may be able to repeat what was heard, but they do not retain or understand it as well as they would in ideal listening conditions.

The primary difference between younger and older listeners with relatively good audiograms (normal pure-tone thresholds below 4 kHz) is simply that older adults need a signal-to-noise ratio (SNR) that is about 3 dB more favorable than that required by younger adults to achieve the same score on a speech in noise test. Older adults' need for a more favorable SNR is due to auditory aging, including age-related differences in auditory temporal processing that are not explained simply in terms of the audiogram (Pichora-Fuller & Souza, 2003; Pichora-Fuller & MacDonald, 2008).

Because SNR conditions that are challenging for an older listener may not be challenging for a younger listener, older listeners may demonstrate the negative effect of effortful listening on cognitive performance in SNR conditions in which younger adults are unaffected. Despite differences among individuals or groups in the specific SNR at which listening becomes effortful, cognitive performance on measures such as memory and comprehension will be reduced for any listener when listening becomes effortful (Rönnberg, Rudner, Foo, & Lunner, in press). Apparent age-related differences in cognitive performance are largely eliminated when younger and older adults are tested in individualized SNR conditions that are equated so that each listener experiences the same degree of listening challenge (Schneider & Pichora-Fuller, 2000).

Physiological Research

In the present era of neuroscience, technological advances have enabled researchers to use tools such as functional magnetic resonance imaging (fMRI) to study patterns of brain activation. These patterns involve connections between different regions of the brain when humans perform complex behaviors, including perception, memory, attention, and language tasks (e.g., Goodglass & Wingfield, 1998). An appreciation of these patterns of activation involving multiple brain regions has superseded the simpler idea that our senses operate in a "bottom-up" modality-specific fashion to relay information from our ears to dedicated regions of auditory cortex (e.g., Scott, 2005; Zatorre, 2007).

In cognitive neuroscience, two common strategies look for differences by comparing patterns of brain activation on two or more tasks or by comparing activation in two or more brain regions to determine the network of connections mediating a task (Horwitz, MacIntosh, Haxby, & Grady, 1995). The first strategy investigates the role of an anatomical site in a variety of tasks; the second strategy investigates how different anatomical sites combine when a task is performed.

Building on these two strategies, "cross-function" and "within-function" approaches have been described in which brain organization is studied to understand how the neural correlates of different cognitive functions overlap and interact (Cabeza & Nyberg, 2002). The goal of the cross-function approach is to determine all of the functions (or tasks) that involve a given brain region; the goal of the within-function approach is to determine all of the brain regions that participate in a given function (or task). These two approaches have been applied to research on adult aging; the research is expected to help explain complex auditory behaviors that rely on a combination of auditory and cognitive processes.

One brain-imaging study showed that the pattern of brain activation shifts when a driver (in a driving simulator) must also listen to speech. This finding is consistent with the idea that cognitive resources are shared among tasks in demanding multi-tasking situations, and that listening consumes cognitive resources that would otherwise be allocated to other tasks (Just, Keller, & Cnyker, 2008). It has been suggested that more widespread brain activation is an indication of "thinking harder" and that it may reflect the allocation of more working memory resources during language processing (e.g., Just, Carpenter, Keller, Eddy, & Thulborn, 1996).

In another brain-imaging study in which young adults with normal hearing listened to sentences in degraded conditions, brain activation was more widespread in more adverse listening conditions (Obleser, Wise, Dresner, & Scott, 2007). Specifically, when sentence context was available, the investigators observed increased activation in the left dorsolateral prefrontal cortical areas that are thought to be involved in semantic processing and working memory (e.g., Petrides, Alivisatos, Meyer, & Evans, 1993). It seems that when listeners engage in more "top-down," context-driven processing of auditory information by involving greater activation of prefrontal cortex, perceptual learning of degraded speech by young adults with normal hearing is accelerated (Davis, Johnsrude, Hervais-Adelman, Taylor, & McGettigan, 2005).

Interestingly, when younger and older adults perform equivalently on various perceptual and cognitive tasks, there is more widespread activation in older brains than in younger brains, with one interpretation being that this difference reflects compensatory processing (e.g., Cabeza, 2002; Cabeza, Anderson, Locantore, & McIntosh, 2002). Such compensatory brain activation would be consistent with the finding that older adults are better than younger adults at using context to compensate for difficulty hearing in challenging listening conditions (Pichora-Fuller & Singh, 2006; Pichora-Fuller, in press). Individual differences in the ability to use context to support the understanding and learning of a degraded or novel signal may be critical to explaining why individuals with higher cognitive function do better with fast-acting, complex-signal-processing hearing aids.

Epidemiological Research

Epidemiological research has uncovered links between age-related changes in audition and cognition that raise serious issues for audiologists to consider as they formulate best practices incorporating cognitive factors. The prevalence of both hearing impairment and cognitive impairment increases with age. Hearing loss is associated with cognitive function, even when the participants are deemed to be clinically normal on cognitive screening tests (Golding et al., 2006). In addition, as many as 90% of people with dementia have hearing loss (Gold, Lightfoot, & Hnath-Chisolm, 1996), and as many as a third of older adults with dementia living in residential care were recategorized as having less severe dementia when they were re-tested with amplification (Weinstein & Amsel, 1986). There is an obvious need to take sensory function into account when assessing cognitive function (Uhlmann et al., 1989b).

Hearing loss may even contribute to or accelerate clinically significant cognitive decline, and it has been suggested that sensory intervention can reduce problem behaviors in dementia (Palmer, Adams, Bourgeois, Durrant, & Ross, 1999), or even slow cognitive decline (Peters, Potter, & Scholer, 1988; Wahl & Heyl, 2003). In one study the relationship between hearing loss and cognitive decline was examined in two groups of 100 people—one group with and the other without Alzheimer's-type dementia—matched with respect to age, sex, and education (Uhlmann et al., 1989a). The investigators found that cognitive function was significantly correlated with hearing loss in both groups, but more so for the group with dementia, suggesting that hearing loss is a risk factor for dementia.

In another large-scale, multi-center longitudinal study of older women who were participating in research on osteoporotic fractures, combined vision and hearing loss was associated with the greatest odds for cognitive decline and for functional decline on five everyday activities over a period of four years (Lin et al., 2004). In a smaller longitudinal study, individuals with Alzheimer's who had hearing loss experienced a more rapid cognitive decline than individuals with Alzheimer's who had relatively good hearing (Peters et al., 1988). Research has also shown a reduced rate of decline in scores on a cognitive screening test over a six-month period following intervention with hearing aids (Allen et al., 2003), suggesting that audiologic rehabilitation may slow the progression of dementia.

Most studies concerning the association between age-related hearing impairment and cognitive impairment have used pure-tone thresholds to measure degree of hearing loss and few have included measures of central auditory processing. In one study, people with probable Alzheimer's disease performed worse on central auditory processing tests, even though they were matched to the control group with respect to age, gender, and pure-tone average (Strouse, Hall, & Burger, 1995). When speech tests were employed to measure central auditory dysfunction, the results were predictive of the likelihood of developing Alzheimer's disease, even after the contribution of audiometric sensitivity had been taken into account (Gates et al., 2002). The epidemiologic research suggests a bidirectional link between auditory and cognitive impairment. Therefore, health professionals assessing cognitive impairment need to know the hearing status of the individuals being assessed; conversely, audiologists need to be aware of individuals' cognitive status when conducting assessments or rehabilitation.

Cognitive Measures

Audiologists will need to find effective ways to participate in interprofessional geriatric teams for older adults who have or are at risk for clinically significant cognitive impairment, so that information about auditory and cognitive function can be shared. For healthy adults, regardless of age, individual differences in cognitive function have emerged as a factor that may help audiologists to understand the candidacy for and/or the outcomes of interventions that aim to improve ease of listening in challenging situations. Two daunting questions remain: What kind of cognitive test would be the best suited for use by audiologists? How would audiologists use cognitive measures?

Much work remains to be done to develop a test of cognitive function that could be used in audiologic practice. However, a working-memory test seems likely to be the most promising for audiologists to use to measure cognition (Pichora-Fuller, 2007). Over the last three decades, the role of working memory in language comprehension has been investigated extensively, so it is reasonable for audiologists to use this existing knowledge and apply it to test development. Indeed, preliminary studies examining alternative tests of cognition have favored working memory over other cognitive measures (Vaughan, Storzbach, & Furukawa, 2006; Foo, Rudner, Rönnberg, & Lunner, 2007). Any test developed will need to be clinically feasible and acceptable to both audiologists and their clients. Adapting existing tests of speech understanding to incorporate working memory may overcome some of the obstacles related to the feasibility and acceptability of the test.

Ultimately, cognitive tests might be used to differentiate individuals with high or low cognitive function and to select interventions taking cognition into account. For example, a person who performs well on a cognitive test might be fitted with a hearing aid that uses fast-acting, complex signal processing; another person who does not perform as well might be fitted with a slower-acting or simpler device or be given training to facilitate acclimatization to a more complex hearing aid.

Another possible application of cognitive testing in audiology would be to evaluate whether easier listening is the result of hearing aid fitting or other types of rehabilitation. For example, for many individuals it is not possible to demonstrate significant differences in benefit between hearing aids using measures such as the scores on standard word discrimination tests. It may be possible, however, to demonstrate differences due to particular hearing aid fittings by measuring how much of what the individual heard was remembered (Edwards, 2007). With recent advances in neuroscience and signal processing technology, we are now poised to finally tackle the issue of how best to incorporate cognitive measures into audiologic practice.  

Acknowledgements: This research has been supported by the Canadian Institutes for Health Research and the Natural Sciences and Engineering Research Council of Canada.