The introduction on timbre really set the tone.
Music research in a scientific context is still in its infancy. Auditory perception beyond mechanical aspects is difficult due to the human element, the metaphorical fixins of our musical cake. While the sensory, bottom-up signal flow of initial oscillation, through the ear, cochlear system, and brainstem is well-mapped, what happens beyond is murky and subjective, relying strongly on the past experiences, preferences, and musical training of the listener (to name a few).
This issue is further confounded by the inconsistent methodology when using musical selections in a research context. Many experiments do not establish a proper control group before extrapolating results, making findings questionable at best. There are at least four levels on which audition occurs, each presenting its own hurdles before achieving rigor. To address all four levels of music experience, research methods should accurately describe all audio samples used not only by providing recordings, but by considering and describing all their aspects as often laid out in an academic musicology setting.
- Attention/distraction/meditation – Is there any differentiation between music with little or no discernible pulse or any extreme features (such as loudness), compared to non-auditory stimulus that occupies similar attention in a neural space?
- Bottom-up/sensory – The level at which we see the most direct neurological effects such as regulating heartrate, breathing, and motor resonance frequencies via music with a strongly discernible pulse. Deals with rhythmicity and low-pitched frequencies, and possibly harmony as experienced pre-analysis in the brain.
- Top-down/cognitive – Short-term memory and responses after the auditory system has provided information to analytical regions of the brain, such as language centers. Related to pitched material especially in vocal ranges, language/lyrical perception, and general musicological analysis.
- Long-term/subjective – Learned responses through nostalgia, cultural, skill/training, and other past experiences that can emphasize or diminish the some or all of these experiential levels.
Thus, the full description of an audio sample will include details of all seven aspects discussed in the previous sections. Tradition, genre, instrumentation, and recording context must all be considered where applicable. For each aspect, an inexhaustive list of example data:
- Rhythm: Tempo/tempi, strength and discernibility, tradition/genre
- Pitch: Melodic, vocal, range, rate of change, language, tradition/genre
- Harmonic: Simple vs. complex (e.g. trance vs. jazz), rate of change, key/mode
- Timbre: Distorted, harsh, soft, mellow, brassy, sparse, reedy, deep, full, etc forever.
- Structure: Short/clip, lengthy/multi-movement, folk, concerto, pop, through-composed, etc.
- Context: Cultural, recording (e.g. live/studio), familiar/unfamiliar, vamping vs. none, general personal taste, etc.
- Loudness: Depends on experimental factors, but can be objectively differentiated such as in metal vs. smooth jazz.
An example description for the audio sample “Beat It” by Michael Jackson might read:
Dance rock studio recording with a strong, steady drumkit beat at 138 BPM. Popular music structure using repetitive melodic phrasing, high male American English vocals, and simple rock harmonies on electric guitar and bass. Extensive vamping, virtuosic instrumental bridge solo.
That is a thorough description of a five minute sample song. Here’s another of Samuel Barber’s “Adagio For Strings” for orchestra:
Eurotraditional symphonic recording focusing on strings, with a weakly discernible, extremely slow molto adagio pulse. Simple harmonic progression in steady arch form, subtle variations in notated meter, large swells in pitch range and volume. Repetitive, slow-moving melody, no vocals, no percussion, no instrumental solos. Often recognizable from film and television.
By using such rigor when selecting music, we can better standardize music description in research environments. Ideally, a paper will include a link to an online recording of audio materials used, as well data from individuals tested regarding personal preferences and tastes before and after listening experiments are conducted.