People go to an audiologist because they have a listening difficulty or they think their child may have one. Listening always involves the following three elements: the ear, central auditory nervous system (CANS; auditory nerve to auditory cortex), and cortical systems beyond the auditory cortex. Consider the simplest act of listening: detecting a pure tone. Neural impulses representing that tone must travel up to the auditory cortex and beyond. Recent research suggests that consciousness of sounds and other sensory stimuli takes place in a nearby cortical region known as the “posterior hot zone” (Nature. 2018 May;557(7704):S8). A person who is insensitive to pure tones has a hearing loss usually due to one or more well-known problems in the ear. But what about someone who has a listening difficulty without a hearing loss? This question has perplexed audiologists, hearing scientists, and others for more than 100 years. An often used term for this condition, “(central) auditory processing disorder” or (C)APD, assumes an origin within the CANS. However, this assumption that the problem resides in the CANS is mostly unfounded, except in cases of neurological disease or trauma. In a recent guest editorial in Ear and Hearing, I explained concerns about the theory and application of (C)APD to childhood listening difficulties (Ear Hear. 2018 Jul/Aug;39(4):617). In this brief essay, I discuss the various origins for what may underlie this “obscure auditory dysfunction” (Ear Hear. 1989 Jun;10(3):200).
As a career-long fan and student of the CANS, I have learned never to ascribe a function to the brain without considering carefully the role of the ear. Several recent discoveries have shown this caution to be well-founded. The most high-profile example has been cochlear synaptopathy (also called hidden hearing loss, HHL1), a loss of ribbon synapses and their associated afferent nerve fibers from the base of inner hair cells (J Neurosci. 2009 Nov 11; 29(45): 1407). Although its impact on human hearing is debated, there seems little doubt that synaptopathy does occur in humans and that its origin and symptoms may be related to a number of other poorly explained phenomena, including APD. Another discovery, also far from fully explored, is extended high frequency (EHF) hearing loss. Several papers searching for synaptopathy have shown that hearing loss in the EHF range (8-20 kHz) is not uncommon in young and otherwise healthy humans. Our own preliminary evidence (Motlagh-Zadeh, et al. Presentation at the American Auditory Society: Scientific & Technology Meeting at Scottsdale, AZ, in February 2018) and another recent paper (Ear Hear. 2018 Jul 26. doi: 10.1097/AUD.0000000000000640) suggest that EHF hearing loss contributes to decreased speech-in-noise perception, a frequent and widely acclaimed symptom of APD. These discoveries point to peripheral origins for some listening difficulties that might be incorrectly ascribed to central APD. Audiologists can begin to address these recent discoveries by measuring EHF hearing in all people complaining of listening difficulty. Synaptopathy does not seem to be easily measurable in humans, but recent animal research on broad-band middle ear muscle reflexes suggests that more sensitive clinical measures are on their way (Hear Res. 2018 Jun;363:109).
COGNITIVE HEARING SCIENCE
Moving to the other end of the system, there has been a dramatic rise over the past 10-20 years of a sub-discipline of hearing research known as cognitive hearing science (Scand J Psychol. 2009 Oct;50(5):371). Cognition has a very broad definition, but the cognitive subdomains recognized by experts as being most important for health and daily functioning are executive function, episodic memory, language, processing speed, working memory, and attention (NIH Toolbox; Neurology. 2013 Mar 12;80(11 Suppl 3):S54). The emergence of cognitive hearing science, with its own journal (Frontiers in Cognitive Auditory Neuroscience) and established international conference, is a testament to the importance we now place on the role of higher-level cortical function in hearing. Although relatively new to hearing science, the close relation between hearing and cognition has long been recognized in cognitive psychology. It has also been hotly debated because it turns out to be extraordinarily difficult to determine whether poor performance on either a hearing or a cognitive task is caused by one or the other. Let's first consider the pure tone audiogram. A listener responding to a quiet tone presentation may miss responses due to limited hearing or lack of attention. Even with close attention, thresholds vary with recent history of tone level (e.g., whether the previous tone was audible) and with practice. As listening tasks become more complex, the cognitive load increases, and performance becomes more erratic and dependent on the cognitive as well as the auditory ability of the listener. There is a considerable cognitive load even for a simple speech-in-noise test such as digits-in-noise (DIN). For example, comparing the performance of individuals on a DIN test and a cognitive test battery in a huge sample of people aged 40-70 (UK Biobank), we found that the most cognitively-able 70-year-olds recognized digits with the same accuracy as the least cognitively-able 40-year-olds (PLoS One. 2014 Sep 17;9(9):e107720). We also examined the relation between audiometric threshold and cognitive performance in the NIH Toolbox database and showed that although tone thresholds were significantly affected by cognitive ability, they were less affected than the DIN.
Consider next the tests most commonly used to diagnose (C)APD in children (Am J Audiol. 2011 Jun;20(1):48). These are staggered spondaic words, various speech-in-noise tests (e.g., SCAN Auditory Figure-Ground), and various dichotic word tests (e.g., dichotic digits). Each of these tests is a complex language-based task that places considerable demand on peripheral and central auditory function and on cognition. How much does performance on these tasks depend on hearing per se, and how much on cognition? Where this distinction has been tested, the data clearly show that cognition dominates. For example, dichotic digits have been very widely used in APD assessment. The results have been interpreted as a measure of binaural integration or binaural separation. However, Cameron and colleagues have shown that if, instead of the normal procedure of presenting different digits simultaneously to each ear, the same digits are presented to each ear (diotic condition), performance in the dichotic condition is highly correlated to performance in the diotic condition (r = 0.8; J Am Acad Audiol. 2016 Jun;27(6):470). These results suggest that while binaural hearing may be disrupted during listening to dichotic digits, it is the ability to listen attentively, separate two simultaneously present sounds, remember what was heard, and recall accurately the heard digits that primarily determines the outcome. These are clearly important and necessary skills needed for healthy listening and academic success that extend far beyond what the ear and CANS alone can provide.
Let's consider a final example from the psychology of aging. To study the relation between cognition and normal hearing, Füllgrabe and colleagues went to great lengths to recruit a sample of older listeners with normal audiograms, at least to 6 kHz, and they then restricted all testing to frequencies less than 6 kHz (Front Aging Neurosci. 2015 Jan 13;6:347). In speech-in-noise testing, they found that carefully matched younger adults performed better than the older listeners. The younger listeners also performed better on the detection of temporal modulation and fine structure and on most (but not all) cognitive tests. Both temporal perception and cognitive performance were correlated with speech identification in noise. The authors of this painstaking and influential study concluded that declining speech perception in the older listeners was caused by cognitive and perceptual impairment separate from audiometric sensitivity. However, modulation masking release, the performance advantage of presenting a target speech sound against a modulated rather than an unmodulated masker (dip listening), did not differ with age. One interpretation is that the perceptual impairment in the older listeners on all the modulation detection tasks was due to cognitive difficulties performing those tasks rather than auditory impairment.
This last example shows how we might separate the cognitive and sensory aspects of hearing using derived or subtractive testing. In this approach, performance on one version of a task is compared with another differing only in one auditory parameter (e.g., masker modulation). Derived testing can be adapted to almost any form of auditory testing (J Commun Disord. 2012 Nov-Dec;45(6):411). The binaural masking level difference (MLD) is another common example of this form of testing. Notably, the two tasks from which the difference is calculated vary only in the interaural phase. It is reasonably assumed these two tasks involve the same attention, memory, and language load. MLD should thus be a sensitive measure of one specific CANS function, provided that the ears are working properly. Several studies have found no change in the MLD in children as a function of age (Ear Hear. 2011 May-Jun;32(3):269) or listening ability/APD (Ear Hear. 2015 Sep-Oct;36(5):527), demonstrating that at least this CANS function (interaural phase comparison) is unaffected by maturity or by APD diagnosed in the recommended way. However, MLD (Arch Otolaryngol Head Neck Surg. 1991 Jul;117(7):718; Audiology. 1991;30(2):91) and spatial hearing (Semin Hear. 2015 Nov;36(4):216) have been found to be reversibly impaired by early middle ear disease. We have found using derived testing that other major aspects of hearing, temporal, and spectral resolutions are largely unaffected by age (see Fig. 1) or cognitive ability (Pediatrics. 2010 Aug;126(2):e382).
FOOD FOR THOUGHT
In conclusion, I propose that most of the developmental immaturity and central impairment in hearing and listening found in children from 6 years of age onwards is due to their inability to attend, understand, and remember challenging auditory tasks rather than an auditory processing disorder. I realize this hypothesis goes against much accepted orthodoxy of developmental science, including some of my own older publications, but it does receive considerable support from studies both inside and outside human hearing science (Int J Psychophysiol. 2015;95(2):125; Atten Percept Psychophys. 2018 Jul;80(5):1311). Concerning maturation, in at least one respect, which is EHF tone detection, children have more sensitive hearing than young adults (Audiology. 1989;28(5):241). And while some younger children are more delayed or disordered in listening skills than their peers, others perform as well as adults (J Acoust Soc Am. 2008 Jun;123(6):4393). There has been much back-and-forth about whether we should retain the term APD and, if so, how it fits alongside listening difficulty. The term “listening difficulty” is nothing more nor less than an umbrella symptom under which specific diagnoses need to be hung. To be truly novel, relevant, and useful for auditory science, a diagnosis of central APD needs to be shown to be a deficit or deficits in the way auditory stimuli are processed that is not due to a problem with the ear or the consequence of a primary deficit in memory, attention, or language. Some close variant of this definition forms the basis of many national statements on APD.
A final note on intervention: For a child who is disproportionately (for age) inattentive to speech, there are a number of options. Behavioral interventions are preferred to other, more invasive treatments. But other than providing instruction on and reinforcing good listening strategies, I don't know of any reliably effective computer-based training. Physical exercise is the most well-evidenced effective intervention for a range of relevant cognitive difficulties (e.g., executive function; Health Psychol. 2011 Jan;30(1):91). Environmental modification can also be useful, of course, if this is something you have control over. Mild amplification or use of remote microphone, personal ear-level systems has been shown effective in several studies (J Am Acad Audiol. 2018 Jul/Aug;29(7):568). Stimulant medication has been found to enhance auditory perception in children (J Speech Lang Hear Res. 2006 Oct;49(5):1072) and other aspects of attention deficits in some children (J Clin Psychiatry. 2018 Mar/Apr;79(2). pii: 17m11553). I would like to see every classroom equipped with a public address (sound field) system, driven by a roving microphone worn by the teacher. That would be a simple, low-cost solution that would help address the communication requirements of every child. While evidence suggests efficacy of such systems (Hearing Journal. 2002;55(3):38), this is an important topic for further research.
Acknowledgements:Current developmental research in our Listening Lab is funded by Cincinnati Children's Hospital Research Foundation and by NIH R01DC014078. Most of our older work cited here was supported by the Medical Research Council (UK). Thanks to the many staff who collaborated in the research and to the families and children who participated. Harvey Dillon and Lisa Hunter and Fan-Gang Zeng provided helpful comments on earlier drafts of this article.
1 The term hidden hearing loss (HHL) has the merit of being a concise and easily understood description of a sub-clinical listening difficulty or one of unknown origin. But many critics correctly point out that it obscures the well-defined mechanism of cochlear synaptopathy.
Thoughts on something you read here? Write to us at [email protected]lterskluwer.com