Can Cortical Auditory Evoked Potentials Demonstrate An Electrophysiological Release From Masking In Noise?

Dr. Frank Musiek
February 6, 2019

Andrew Stuart1 & Sarah P. Faucette2
1Department of Communication Sciences & Disorders, East Carolina University, Greenville, NC
2Department of Otolaryngology & Communicative Sciences, The University of Mississippi School of Medicine, Jackson, MS


Clinicians typically evaluate auditory function with behavioral testing. When patients cannot be assessed with behavioral techniques, auditory evoked potentials are often employed. Auditory evoked potentials provide “objective” information regarding the structure and function of the auditory neurological pathway. Specifically, auditory evoked potentials have been used to estimate behavioral audiometric thresholds; evaluate auditory development; assess the assistance from cochlear implantation, auditory training, or amplification; and examine auditory discrimination and speech perception.

We have been interested for some time in the ability of listeners to better understand speech in background competitors that fluctuate in level, intermittency, or both level and intermittency relative to stationary competitors (e.g., Stuart, 2008; Stuart & Phillips, 1996; Stuart et al., 1995, 2006, 2010). Two general paradigms have been used to demonstrate this perceptual advantage – what we refer to as a “release from masking” (RFM). In one paradigm, speech stimuli are fixed in level while continuous and interrupted broadband noises are presented at equal and varying signal-to-noise ratios (SNRs; i.e., fixed speech). In the other paradigm, the noises are fixed in level and thresholds for speech stimuli are determined (i.e., fixed noise).

Listeners have superior speech recognition scores and lower speech reception thresholds in interrupted compared to continuous noise. This performance enhancement has been attributed to the auditory system’s temporal ability to resolve speech fragments, or get “glimpses” or “looks” of speech, between the gaps of noise. This temporal resolution ability allows the listener to patch the information together to identify the specific speech stimuli. We have argued that listeners who fail to display an advantage in the interrupted noise have a deficit in temporal resolution when compared to normal listeners. More recently, we have been interested in whether this same RMF can be demonstrated with cortical auditory evoked potentials (CAEPs). Demonstrating an eletrophysiological measure of RFM could lead to better understanding of the underlying mechanisms of temporal resolution, as well as serve as an objective measure of the temporal component of speech in noise understanding.

It has been recognized that CAEPs can be used to study the encoding of stimuli in the presence of stationary noise. Background noise can affect the passively evoked P1-N1-P2 components of the CAEP response. For example, CAEP responses to speech and tonal stimuli, in the presence of stationary noise, have reduced amplitude and prolonged latencies compared to responses in quiet (Billings et al., 2009, 2011). Decreases in SNR also cause longer latencies and smaller amplitudes (Billings et al., 2013). In some conditions, however, CAEPs are enhanced. N1 is enhanced in low-level background noise, relative to quiet, and P1 and P2 components are larger with binaural versus monaural presentation (Papesh et al., 2015).

Few researchers have examined the CAEP in the presence of non-stationary background competitors. Androulidakis and Jones (2006) recorded N1 and P2 responses evoked to a 1000 Hz tone in quiet and in the presence of unmodulated and modulated noise. N1 and P2 were only observed in the unmodulated noise condition. Billings et al. (2011) recorded CAEPs with a tonal and speech stimulus in continuous speech spectrum noise, interrupted speech spectrum noise, and four-talker babble. The stimuli were presented in quiet and with the noises at -3 dB SNR. P1 and N1 latencies were significantly prolonged and N1 amplitudes were reduced in all three noises relative to quiet. Billings et al. did not find a physiological RFM. CAEP response differences between interrupted and continuous noise were minimal.

We recently reexamine CAEPs to a speech stimulus in continuous and interrupted competing noises (Faucette & Stuart, 2017) with young adult listeners. The two noises had identical spectra and differed only in their temporal structure (i.e., the interrupted noise had rectangular on/off envelops with noise bursts and silent periods with durations of both varying randomly from 5 to 95 ms with a duty cycle of 0.50). CAEPs were evoked with a speech token /da/ presented in quiet and in competing continuous and interrupted noises. Two paradigms that mimic previous behavioral studies by Stuart and colleagues were utilized. A fixed speech paradigm used a constant 65 dB SPL speech presentation level and the noises varied (i.e., +10, 0, and -10 dB SNR). A fixed noise paradigm used a constant 65 dB SPL background level and speech stimulus varied. Beginning at 0 dB SNR, SNR was reduced by 10 dB until no response was present. The SNR was then increased in increments of 5 dB until response was once again measurable. “CAEP SNR threshold” was defined as the lowest SNR with a measurable response.

Grand average waveforms of CAEPs in quiet and as a function of noise and SNR are also shown in Figure 1. Clearly, in the fixed speech paradigm, one can see that as SNR decreased, latencies were prolonged and peak-to-peak amplitudes decreased – with the effect more pronounced in continuous noise. RFM is evident in the -10 dB SNR condition as the difference between waveforms. We also found that only 28% of participants demonstrated a CAEP at the poorest -10 dB SNR. RFM was also evident in the fixed noise paradigm: CAEP SNR thresholds were significantly lower in the interrupted noise (M = -24.2 vs. M = -7.2 in continuous noise).

We find these electrophysiological results exciting as they parallel the pattern of behavioral performance with measures of speech recognition in varying SNR and sentence reception thresholds in both continuous and interrupted noise (Stuart & Phillips, 1996; Stuart et al., 1995, 2006). We caution, however, any clinical application of these CAEP measures remains to be established. That is, we have not at this time demonstrated a relationship between electrophysiological and behavioral measures of RFM in the same group of listeners. There have been previous reports of CAEPs significantly predicting speech perception in noise with children (Anderson et al., 2010) and adults (Billings et al., 2013). Will the RFM observed with these CAEPs be related to RFM measures observed with behavioral speech recognition in noise? We are currently examining the relationship between RFM electrophysiological and behavioral measures in the same cohorts of young and old normal hearing and hearing impaired listeners. If a relationship between electrophysiological and behavioral measures of RFM CAEPs exists it may lead to a better understanding of clinical complaints of speech in noise difficulties.


  1. Anderson, S., Chandrasekaran, B., Yi, H.G., & Kraus, N. (2010). Cortical-evoked potentials reflect speech-in-noise perception in children. Eur J Neurosci, 32, 1407-1413.
  2. Androulidakis, A., and Jones, S. (2006). Detection of signals in modulated and unmodulatednoise observed using auditory evoked potentials. Clin Neurophys, 117, 1783-1793.
  3. Billings, C.J., McMillan, G.P., Penman, T.M., & Gille, S.M. (2013). Predicting perception in noise using cortical auditory evoked potentials. JARO, 14, 891-903.
  4. Billings, C.J., Bennett, K.O., Molis, M.R., & Leek, M.R. (2011). Cortical encoding of signals in noise: Effects of stimulus type and recording paradigm. Ear Hear, 32, 53-60.
  5. Billings, C.J., Tremblay, K.L., Stecker, G.C., & Tolin, W.M. (2009). Human evoked cortical activity to signal-to-noise ratio and absolute signal level. Hear Res, 254, 15-24.
  6. Faucette, S.P., & Stuart, A. (2017). Evidence of a speech evoked electrophysiological release from masking in noise. J Acoust Soc Am, 142, EL218-EL223.
  7. Papesh, M.A., Billings, C.J., & Baltzell, L.S. (2015). Background noise can enhance cortical auditory evoked potentials under certain conditions. Clin Neurophysiol, 126, 1319-1330.
  8. Stuart, A. (2008). Reception thresholds for sentences in quiet, continuous noise, and interrupted noise in school-age children. J Am Acad Audiol, 19, 135-146.
  9. Stuart, A., and Phillips, D.P. (1996). Word recognition in continuous and interrupted broadband noise by young normal-hearing, older normal-hearing, and presbyacusic listeners. Ear Hear, 17, 478-489.
  10. Stuart, A., Givens, G.D., Walker, L.J., & Elangovan, S. (2006). Auditory temporal resolution in normal hearing preschool children revealed by word recognition in continuous and interrupted noise. J Acoust Soc Am, 119, 1946-1949.
  11. Stuart, A., Phillips, D.P., and Green, W.B. (1995). Word recognition performance in continuous and interrupted broad-band noise by normal-hearing and simulated hearing-impaired listeners. Am J Otol, 16, 658-663.
  12. Stuart, A., Zhang, J., & Swink, S. (2010). Reception thresholds for sentences in quiet and noise for monolingual English and bilingual Mandarin-English listeners. J Am Acad Audiol, 21, 239-248.

Leave a Reply