Unmasking Auditory Temporal Masking

Dr. Frank Musiek
January 7, 2015

By Renata Filippini, Ph.D.
University of Sao Paulo

Auditory temporal masking may be didactically considered an auditory temporal processing ability, alongside temporal ordering, temporal resolution and temporal integration (Shinn, 2003). Currently, clinical tests for the assessment of temporal ordering and temporal resolution are available to clinicians, while tests to evaluate temporal masking and temporal integration were not yet developed. Perhaps, for that reason, the concept of temporal masking is not as clear to the clinical audiologist as the concept of temporal resolution and temporal ordering, and even of clinical masking.

In temporal masking, sounds presented in sequence, non-simultaneously, influence each other’s perception, as opposed to clinical masking, in which a sound influences the perception of another sound simultaneously presented. Auditory temporal masking can occur in two situations: when a signal masks a subsequently presented signal (forward masking), and when a signal masks a previously presented signal (backward masking) (Figure 1).

On one hand, while performing a task of clinical masking, because they are presented simultaneously, one should be able to perceive the frequencies of the target signal among the frequencies of the masker. On the other hand, while performing a task of temporal masking, to realize that two signals are being presented, one must be able to perceive the temporal cues between them. Therefore, in the former people rely mostly on frequency resolution, and in the latter people rely mostly on temporal resolution (Hartley et al., 2000).

Diagram representing the three situations auditory masking can occur.

Diagram representing the three situations auditory masking can occur.

Although the study of auditory temporal masking dates back to the 1950’s and 60’s, researchers still do not fully comprehend the mechanisms involved in forward and backward masking. Some of the hypotheses propose: (1) the integration of both sounds, which are then processed as only one stimulus; (2) the interruption or inhibition of the neural processing of one signal by the processing of the other signal; (3) differences in the transmission velocities of each sound information, so that even if they are presented in distinct moments, both would reach the superior levels of the auditory system at the same time, then being analyzed as just one signal, and, finally, (4) the association of two or more of the above hypotheses (Kallman et al., 1986; Crawley et al., 1994; Durrant and Lovrinic, 1995; Shin, 2014).

Some researchers believe that forward and backward masking must have different mechanisms, the first being primarily peripheral and the second primarily central. Such difference is assumed because backward masking has shown more influence from cognitive factors, such as attention and memory, and subjects present much better results after training (Hartley and Moore, 2002; Boets, 2011). The fact is forward masking is the better understood of the two, keeping great similarity to other auditory properties like adaptation and fatigue (Moore, 2013). Thus, most of what we know so far is more easily applied to forward than to backward masking.

What we do know is that auditory temporal masking occurrence depends on several characteristics of both sounds (masker and masked), such as their frequency and duration, as well as their intensity, and the interval between them.    Two stimuli with similar frequency spectrums show higher probability for the occurrence of temporal masking, both forward and backward (Gelfand, 2009). If two sounds differ too much regarding that parameter normal hearing and normally developed people will be able to use frequency cues to detect both sounds, especially if they are separated in time.

The duration of the masker seems to influence the occurrence of forward but not of backward masking (Elliot, 1967). Masker sounds with duration from 20 up to 200 ms demonstrate increasingly masking effects, with no increase of the effect by longer maskers (Shinn, 2014). The influence of the masker’s level also seems to play a different role for backward and forward masking. For backward masking, higher intensities do not seem to increase the masking effect like it does for forward masking, in which the increase in masking effectiveness is not linearly equivalent to the increase in intensity: 10 dB increase in masker level result in 3 dB increase in masker effect (Hartley and Moore, 2002; Moore, 2013).

Finally, the increase of the interval between the two signals decreases the masking effects. For backward masking, a more abrupt reduction in the masking effect is produced increasing the interval up to 25 ms, with less reduction for larger intervals, and no effect at all around 100 ms (Shinn, 2014). For forward masking, the masking effect is larger with intervals up to 50 ms, and less evident influence of the masker may be observed with intervals up to 200 ms (Moore, 2013). Besides that, it has been stated that temporal masking is more effective when the masker is presented after the masked signal (backward) and when both stimuli are presented monotically (Eliott, 1962).

But, why should we be as acquainted to temporal masking as we are to the clinical masking?

In complex stimuli, like speech and music, the acoustic information is presented in a brief and sequenced manner, allowing the temporal masking phenomena to take place. Thus, there are some acoustic features of the sounds that even normal hearing people might not perceive, with no loss of information for the comprehension of the message. This happens so often during complex sound perception that is availed in digital compression of audio files like MP3, in which every sound that cannot be perceived by the human ear is eliminated in order to make the file smaller. However, people with auditory processing deficits may also have trouble in perceiving key acoustic features for decoding the message, influencing the perception of speech sounds and even auditory and language development in children.

In 1997 Wright and colleagues reported, for the first time, poor performance in individuals with language disorders in a backward masking paradigm. That study gave a start to a series of other studies comparing the responses of typically developed individuals with those of individuals presenting with Specific language disorder, Dyslexia, and Reading Disorders, aiming to investigate the relationship between temporal masking deficits and language disorders. It is not hard to understand why researchers would hypothesize such a relationship, especially when the role of acoustic temporal cues in identifying and discriminating syllables and phonemes, and the potential for the occurrence of temporal masking between phonemes in natural speech is well recognized (Shinn, 2014). Figure 2 demonstrates how the occurrence of the backward masking effect could blur or eliminate important cues that help with acoustic discrimination of any two syllables. In this example, the syllables are differentiated from each other essentially by the initial acoustic information; if the sounds that follow mask the initial sounds, the syllables might ultimately be perceived as the same.

Representation of how temporal masking could affect the perception of key acoustic elements for sounds discrimination.

Representation of how temporal masking could affect the perception of key acoustic elements for sounds discrimination.

Some of the studies proposed after that first report from Wright, also demonstrated that individuals with language disorders have poorer performance, especially, to backward masking tasks (Marler et al., 2002; Montgomery et al., 2005; Howell et al., 2000). However, other studies failed in observing such differences (Bishop et al., 1999), or did observe them, but with overlapping results (Rosen and Manganari, 2001), or with no correlation at all between performance in language assessment and auditory temporal masking tasks (Rosen et al., 2009).

The association of auditory temporal masking deficits and language disorders is not yet clear or free of controversies. These controversies arise mainly from the high variability observed in many of these studies, as well as the use of distinct methodologies, and the difficulties in separating cognitive and sensory contributions in the tasks. This is probably why a clinical test to evaluate temporal masking have not yet been fully developed. The looming challenge is how to standardize an expected performance to diagnose dysfunction, when performance from normal individuals vary greatly.

The paradigm normally applied in these studies use psychoacoustic adaptive procedures, largely based in manipulation of the masked signal level to find the intensity threshold for its detection. Would intensity be the best feature of sound to be manipulated in such assessment? Given the importance of the interval between stimuli in temporal masking phenomenon, would this be a better parameter? Or would analyzing interval thresholds only provide results with the same variability, because this variability could be more related to the cognitive requirements of the task than to the sensory requirements? Is it possible to develop a temporal masking test that does not rely so strongly in cognitive abilities?

This and other questions have to be answered in the near future. The development of a clinical test to evaluate temporal masking will allow audiologists and researchers to assess auditory temporal processing more thoroughly. Standardizing the procedure and enlarging the available data on temporal masking, would also help the comprehension of temporal masking mechanisms, and could enable us to solve some of the controversies regarding the relationship between language and auditory processing disorders. Considering the aforementioned, this field definitely deserves further attention from researchers and clinicians.

Renata Filippini has a PhD in Rehabilitation Sciences from the University of Sao Paulo (USP), Brazil. In her thesis she studied the effects of auditory training in speech perception of children presenting language disorders, through the use of ABRs to complex sounds. Her interest in this field led her to deepen her knowledge on the temporal aspects of hearing perception. Since the beginning of 2013 she is a post-doctorate researcher at the Auditory Processing Laboratory coordinated by Dr. Eliane Schochat in the University of Sao Paulo. Dr. Filippini’s current study involves auditory temporal masking, and is funded by the Sao Paulo Research Foundation – FAPESP.”


  1. Bishop, DVM, Carlyon RP, Deeks JM, Bishop SJ. Auditory temporal processing impairment: Neither necessary nor sufficient for causing language impairment. Journal of Speech, Language, and Hearing Research. 1999; 42: 844-9.
  2. Boets B, Vandermosten M, Poelmans H, Luts H, Wouters J, Ghesquiere P. Preschool impairments in auditory processing and speech perception uniquely predict future reading problems. Research in Developmental Disabilities. 2011; 32: 560-70.
  3. Crawley EJ, Kallman HJ, Neely JH. A test of an interruption/temporal-uncertainty of auditory backward recognition masking of target duration. Acta Psychologica.1994; 87: 1-18.
  4. Durrant JD, Lovrinic JH. Introduction to psychoacoustics: Temporal Aspects of Hearing. In.: Durrant JD, Lovrinic JH. Bases of Hearing Science. Lippincott Williams & Wilkins. 1995; 294-9.
  5. Elliot, L. Backward masking: Monotic and dichotic conditions. Journal of the Acoustical Society of America. 1962; 34: 1108-15.
  6. Elliot, L. Development of narrow-band frequency contours. Journal of the Acoustical Society of America. 1967; 13: 93-117.
  7. Gelfand, SA. Hearing: An introduction to psychological and physiological acoustics. 5th ed. London, UK: Informa Healthcare; 2009.
  8. Hartley DEH, Wright BA, Hogan SC, Moore DR. Age-Related Improvements in Auditory Backward and Simultaneous Masking in 6- to 10-Year-Old Children et al. Journal of Speech, Language, and Hearing Research. 2000; 43: 1402-15.
  9. Hartley DEH, Moore DR. Auditory processing efficiency deficits in children with developmental language impairments. Journal of Acoustic Society of America. 2002; 112(6): 2962-6.
  10. Howell P, Rosen S, Hannigan G, Rustin L. Auditory backward-masking performance by children who stutter and its relation to dysfluency rate. Perceptual und Motor Skills. 2000; 90: 355-63.
  11. Kallman HJ, Hirtle SC, Davidson S. Recognition masking of auditory duration. Perception and psychophysics. 1986; 40: 45-52.
  12. Marler JA, Champlin CA, Gillam RB. Auditory memory for backward masking signals in children with language impairment. Psychophysiology. 2002; 39: 767-80.
  13. Montgomery CR, Morris RD, Sevcik RA, Clarkson MG. Auditory backward masking deficits in children with reading disabilities. Brain and Language. 2005; 95: 450-6.
  14. Moore BCJ. An introduction to the psychology of hearing. 6. Ed. San Diego: Academic Press; 2013.
  15. Rosen S, Manganari E. Is there a relationship between speech and nonspeech auditory processing in children with dyslexia? Journal of Speech, Language, and Hearing Research. 2001; 44: 720-36.
  16. Rosen S, Adlard A, der Lely, HKJ. Backward and Simultaneous Masking in Children with grammatical specific language impairment no simple link between auditory and language abilities. Journal of Speech, Language, and Hearing Research. 2009; 52: 396-411.
  17. Shinn JB. Temporal processing: the basics. Hear J. 2003; 56(7): 52.
  18. Shinn JB. Temporal Processing Tests. In.: Musiek FE, Chermak GD. Handbook of Central Auditory Processing Disorder – Volume 1. 2nd ed. San Diego: Plural Publishing. 2014. 405-34.
  19. Wright BA, Lombardino LJ, King WM, Puranik CS, Leonard CM, Merzenich MM. Deficits in auditory temporal and spectral resolution in language-impaired children. Nature. 1997 May 8; 387 (6629): 176-8.




birdsong hearing benefits

Leave a Reply