How do we measure musical expectations?

There are at least 8 different experimental methods to characterize a listener’s expectations.

1: Method of tone detection:

One of the earliest methods for tone detection was used by Gordon Greenberg and Willard Larkin in the1960s. They had participants listen to tones in the presence of continuous loud noise. The listeners’ task was to indicate whether or not they heard the tone.

Greenberg and Larkin discovered that listeners were better able to detect a tone if they expected a tone of a specific pitch to occur at a particular moment.

The method of tone detection is rarely used today.  But the work of Greenberg and Larkin remains important as it demonstrates two general principles concerning expectation. First, accurate expectation facilitates perception. When the events of the world conform to our expectations, we are better able to detect, perceive, and process these events. Second, that low-level sensory processes (like the hearing threshold for detecting a tone) are influenced by higher-level mental processes (like expectation).

2: Method of production

James Carlsen, Pierre Divenyi, and Jack Taylor played a sequence of tones and asked listeners to sing what they thought would be an appropriate continuation. Carlsen used this method to compare the melodic continuations of American, German, and Hungarian listeners.6 In analyzing the sung continuations, Carlsen found significant differences between the three groups, suggesting that one’s cultural background influences listener expectations of what might happen next.

This method has several disadvantages, as it requires that participants have some singing ability (and be willing to sing while being recorded).The method also relies on the participants’ facility and comfort with improvising.

3: Prone-tone method

The best-known experimental method for testing musical expectations is the probe-tone method pioneered by Roger Shepard and Carol Krumhansl.

A musical context is presented, like several chords or the initial notes of a melody. After this, a single tone or chord is played, and the listener is asked to judge this target (or “probe”) sound according to some criterion. Often, the listener is asked to judge how well the tone or chord “fits” with the preceding musical context. The original contextual passage is then repeated and a different probe tone or chord is played. Following each presentation, the listener is asked to judge how well the new tone or chord fits with the preceding context.

An advantage of the probe-tone method is that a detailed picture can be assembled where the listener provides information concerning several possible continuations, rather than only a single continuation.

4:  Betting paradigm

In the betting paradigm, participants are given a “grub stake” of poker chips and asked to place bets on a set of possible continuations. Participants hear an antecedent passage and are invited to bet on what pitch they think will occur next.

A benefit of the betting paradigm is that it allows the experimenter to calculate the subjective probabilities for different continuations. Bets should be placed in proportion to the subjective likelihood of subsequent events. For example, if a participant thinks that a certain pitch is twice as likely as another pitch, then the participant ought to place twice as large a bet on the more probable pitch.

5: Head-turning paradigm

When we hear an unexpected sound, we will often turn our head in the direction of the sound. This basic reflex is referred to as the orienting response.

If a stimulus is repeated, after a while an individual will habituate to the stimulus and fail to orient to it. Further repetitions are unlikely to provoke a response. If a change is then made to the stimulus, and if the change is sufficiently novel, then a listener might reorient to the sound. This reorienting to a modified stimulus is called dishabituation. If an infant reorients to a modified sound, then one might interpret this as evidence that the infant didn’t expect the sound.

6: Bradycardic Response Method

Another version of the dishabituation paradigm examines changes of heart rate rather than head movements.

When something catches our attention or is unexpected, our heart rate reduces for a bit and then recovers to its natural rate. This response is referred to as bradycardia.

7: Reaction Time Method

When you hear an expected sound, you will typically be able to process it more quickly and respond to it faster (if a motor response is required). A quick reaction time is therefore correlated with high expectation.

Aarden concocted a task where listeners were required to process and respond to sounds as quickly as possible. While they listened to an ongoing melody, Aarden asked his listeners to indicate whether the pitch contour of the melody had ascended, descended, or remained the same. The listener must press one of three keys as quickly as possible after each note in the sequence. The responses are collected, including the elapsed time between the onset of the heard tone and the key press. The method is based on the assumption that if the pitch contour of a note moves as expected, then this will have a facilitating effect and so produce faster reaction times. Conversely, if a tone moves to an unexpected tone, this will increase the processing time and so result in a slower reaction time.

A drawback of this method is that when a listener makes a slow response, we have no idea of why this occurs—except that the heard tone evoked a more time-consuming mental process.

Method 8: Evoked Response Potential (ERP)

The activity of neurons results in tiny electrical currents. When large numbers of neurons are active at the same time, the aggregate electrical current can often be detected through the scalp using suitably sensitive electrodes.

The most pertinent research related to expectation involves those electrical patterns that arise in response to a particular stimulus, like a tone. Since the recorded brain activity is in response to a stimulus, the ensuing electrical behavior is referred to as an evoked response potential or ERP.

Like the head-turning paradigm and the bradycardic response method, ERP can be useful for studying the expectations of nonhuman animals, and especially for studying preverbal infants.

Based on the article: Zingaro, V. (2016). David Huron, Sweet anticipation: Music and the psychology of expectation. Aisthesis. Pratiche, Linguaggi E Saperi Dell Estetico, 9(1), 201