The phenomena of consonance and dissonance have attracted considerable interest among music scholars throughout history. Although musical consonance is known to be influenced by social, cultural, and other learned factors, response patterns continue to show transcultural similarities that suggest fundamental physiological concomitants.

The stable or invariant core of these responses raise an important question: Why do humans find certain sound stimuli "ugly" or "repulsive?" Why would some simple sine-tone combinations cause distress or discomfort?

This paper reviews the experimental literatures pertaining to sensory or "tonal" dissonance and notes the close links with auditory masking. The relationship between emotions and evolutionary adaptations are noted and Dimberg's (1989) physiological measures of a possible "auditory disgust response" are reviewed.

A speculative theory is then proposed. It is suggested that auditory dissonance is a negative-valence emotion that arises in response to stimulus-engenedered degradation of the auditory system. In short, sounds that are recognized as reducing our capacity to hear other sounds tend to evoke an unpleasant phenomenal experience which in turn leads to stimulus-aversive behaviors. Work by Zuckerman (1994) has linked individual differences in evoked cortical potentials with personality meausres of impulsive sensation-seeking. Similarly, Kimmel et al (1979) showed differences in auditory-evoked heart-rate responses that bifurcate according to sensation-seeking personality correlates. These studies suggest physiological factors that might account for the observed differences in individual responses to dissonance. Problems with this theory are then discussed. Sound examples will be served.

• Biological functions evolve as adaptive survival mechanisms. (Darwin, 1854)
  
  
• Why does hearing exist?
  
  Because the auditory system enhances survival. It provides a source of information about the environment; it alerts us to dangers and opportunities, and allows us to communicate that information.

• For any common behavior it is fruitful to ask how the behavior might enhance survival.
  
  
• Behaviors that involve strong feelings of pleasure or displeasure are thought to be particularly strongly linked to survival.
  
  e.g., food, fear, sex, disgust
  
  
• Brain mechanisms that evoke pleasurable and unpleasurable states may be regarded as evolutionarily adaptive systems that provide the basis for negative and positive behavioral reinforcements.

1. "basic emotion" -- evident from infancy
   
   
2. partial closure of the nose due to flexion of the corrugator muscles
   
   
3. extension of the tongue
   
   
4. backward movement of the head (stimulus aversion)
   
   
5. spitting if stimulus is gustatory.

• Response can be seen in face of someone viewing a photograph, hearing or relaying a story, being told the price of an object..
  
  
• Tongue thrusting as taunt.
  
  
• Laver (1980) has noted that corrugator activity characterizes the facial "sneer" and that due to the physical configuration, concurrent vocalizations exhibit accented nasals (such as the vocalized "nya").

• Can sound/music evoke pleasure?
  Yes. (Goldstein, 1980)
  
  
• Can sound/music evoke displeasure?
  Yes. (Dimberg, 1989)
  
  
  
  
• Is music-making an adaptive or non-adaptive form of pleasure-seeking?

• sounds associated with injury, illness, or pain (such as moaning, weeping, or gagging)
  
  
• unwanted sounds (such as sounds that disrupt sleep or mental concentration)
  
  
• sounds that have a learned negative connotation or association
  
  
• sounds that interfere with other sounds of interest
  
  
• other ...

• Very long and intricate history of research and speculation.
  (Historical surveys: Cazden, 1945, 1980; Tenney, 1988; Greenwood, 1990; Keislar, 1991)
  
  
• The terms "consonance" and "dissonance" are used to identify an extraordinarily wide variety of phenomena.
  
  
• Cultural differences clearly exist.
  
  
• Within individual cultures, musicians are known to respond differently than non-musicians (not just more reliable).
  
  
• Dissonance is not merely the absence of consonance (or vice versa).

• a raw sensory unpleasantness that arises from the interaction of concurrent sound stimuli (called "tonal consonance" by Plomp & Levelt, 1965; called "sensory dissonance" by Kameoka & Kuriyagawa, 1969; now called "auditory dissonance")
  
  
• a thwarting or delaying of expectation that creates a sense of tension or apprehension
  
  
• some types of auditory surprise

• Some concurrent sound combinations are commonly described as more "pleasant," "euphonious," or "harmonious" than other sound combinations. E.g.
  
  N.B. The descriptive adjectives people apply to these sounds are explicitly evaluative rather than neutral.
  
  N.N.B. Although there are some cultural differences in the perception of consonance, there are cross-cultural universals. In evaluating frequency dyads, there are strong cross-cultural correlations in responses. (Kameoka & Kuriyagawa, 1969a/1969b)

• Simple integer ratios. (Zarlino, 1558).
  
  

  PROBLEMS: Diminished triad (5:6:7) doesn't sound more consonant than minor triad (6:7:9). Intervals in low register are more dissonant than same interval in high register.

  
  
  
• Tonal fusion. (Euclid, ~600 BC; Stumpf, 1898).
  
  

  PROBLEMS: Adding harmonics increases fusion but also increases dissonance. Direct judgements of tonal fusion do not match judgements of consonance.

• Synchrony of neural firings. (Boomsliter & Creel, 1961).
  
  
• Length of periodicity. (Irvine, 1946).
  
  
• Time required for pitch perception. (Resnick, 1981).
  
  
• Pattern-matching of frequency template. (Terhardt, 1974).

• Beating/roughness. (Helmholtz, 1877).
  
  

  PROBLEMS: Two intervals that generate the same beat frequency will differ in perceived consonance. For fusion-controlled stimuli, beating does not correlate with dissonance judgments (Keislar, 1991).

  
  
  
• Place Theory. (Greenwood, 1961, 1991).
  
  
• Plomp & Levelt (1965).
  
  
• Dissonance also increases with sound pressure level (Kameoka & Kuriyagawa, 1969a/1969b).
  
  
• Replicated by Nordmark & Fahl\o'e\(aa'n (1988).
  
  
• Good correlates of auditory dissonance found in cochlear models (Simpson, 1994).

• Effect of critical-band evident in musical practice (Huron & Sellmer, 1992).
  
  
• Effect of tonal fusion independent of sensory consonance in musical practice (Huron, 1991).
  
  
• Interval-class content in equally-tempered pitch-class sets: common scales exhibit optimum tonal consonance (Huron, 1994).

• Masking is the tendency for one sound to obscure another.
  
  
• In masking, the presence of one tone in effect raises the threshold of audibility for neighboring tones.

When two tones are roughly similar in amplitude, masking may be complete when they are quite close in frequency.

• Masking degrades the auditory system.
  
  
• i.e., masking reduces our capacity to detect danger or opportunity.
  
  
• i.e., masking is a potential threat to survival.
  
  
• i.e., masking is bad.
  
  
• Masking may be an inevitable limitation of the peripheral auditory system.

• Recruit other auditory processes to circumvent masking. E.g. Comodulation masking release and auditory induction (Thurlow, 1957; Houtgast, 1971; Warren, Obusek & Ackroff, 1972)
  
  
• Avoid acoustic situations where masking occurs.

That is, encourage behavioral aversion to stimulus-induced auditory degradation.

HOW? Evolve a mechanism so that stimuli that degrade auditory performance are perceived as ugly.

• When masking is absent (i.e., when tones are separated by more than a critical band) there would be no need to evoke an aversive response.
  
  
• When masking is complete, a maximum aversive response would be appropriate. However, when masking is complete, the auditory system has no direct way of knowing that masking is occurring.
  
  
• When masking is incomplete, then (1) it is possible for the auditory system to infer that masking is occurring, and (2) it is appropriate to generate an aversive response.
  
  
• At some mid-point (say half the distance of a critical band), two pure tones must necessarily engender significant auditory masking, yet are more likely to remain detectable as separate tones. That is, the two tones not only interfere with each other, but they are perceived as interfering with each other.
  
  
• Accordingly, we might expect a maximum negative valence emotion to arise when pure tones are separated by roughly half a critical band (i.e. about 0.5 mm).

1. Why do people like waterfalls?
   
   
2. "But I like rock 'n' roll" ...

• History of Western music is one of apparent continuous increase in auditory dissonance.
  
  
• Dissonance in rock 'n' roll: (1) increased loudness, (2) use of harmonic distortion, (3) higher energy levels in the bass.
  
  
• With our technological sophistication and knowledge, why hasn't music evolved toward less and less auditory dissonance?

1. Habituation. Exposures occur in safe environments.
   
   
2. Thrill-seeking behavior. Dissonance may be a musical spice.

• Within ~4 seconds of hearing an unexpected tone, a listener's heart-rate typically: (1) increases, or (2) decreases and then increases.
  
  
• Heart-rate acceleration is indicative of a startle response or defense reflex.
  
  
• Heart-rate deceleration-acceleration typifies interest or openness to the stimulus (Graham, 1979).
  
  
• Heart-rate deceleration-acceleration responses tend to occur most commonly for those listeners who score high on "sensation-seeking" personality characteristics (Orlebeke & Feij, 1979; Ridgeway & Hare, 1981)
  
  
• Similar personality-linked differences have been observed in auditory evoked potentials -- see Zuckerman (1994).

• Sensory systems evolved as adaptive mechanisms that enhance survival.
  
  
• Masking reduces the ability of the auditory system to detect all sounds present in the environment. i.e. Masking degrades the auditory system.
  
  
• Hence masking has the potential to reduce survival.
  
  
• Any behavioral tendency to avoid sounds that generate high levels of masking might be expected to confer an evolutionary advantage.
  
  
• A suitable way of encouraging such avoidance behavior would be to have the stimulus evoke a negative valence emotion.

• How is it that some very simple sound stimuli (such as pairs of pure tones) are able to evoke negative phenomenal experiences?
  
  
• Why are sounds that are dubbed "dissonant" experienced as relatively less pleasant than consonance? Why isn't it the reverse -- with nominally "consonant" sonorities perceived as more unpleasant?
  
  
• Why is dissonance proportional to sound pressure level?
  
  
• Why is dissonance related to the critical band?