Listening test stimuli often differ across multiple unknown perceptual attributes simultaneously

Evidence

The previous sections discussed listening tests, where it is often implicitly assumed that the difference between auditory events is in a single auditory attribute, such as in loudness or pitch. As will be discussed later in the context of sound quality, the auditory events studied may differ from each other in multiple dimensions, for example in both loudness and pitch. The situation becomes even more challenging if the attributes are not known a priori. For example, when perceptually motivated audio codecs are developed, and different versions of the codecs are tested, the listeners may perceive changes in ‘crispness’, ‘noisiness’, and ‘loudness’. Descriptive sensory analysis is a family of methods which targets revealing the palette of attributes of a given set of stimuli, and in some cases also scaling (or grading) the stimuli in the attribute dimensions. — Ville Pulkki, Matti Karjalainen, Communication Acoustics- An…, 930

Categorization

Category: Definitional Content

• DEFINE: Introduces descriptive sensory analysis as a method for uncovering unknown perceptual attribute palettes in stimuli.
• XREF: Connects to audio codec dev/testing methodology and to sensory evaluation methods used in food and perfume industries.

Related

• Audio codec testing requires methods that reveal unknown perceptual attributes (similarity 0.95)
• Descriptive sensory analysis maps perceived sound differences onto an N-dimensional attribute space (similarity 0.87)
• Careful listening tests validly reveal how acoustic attributes become auditory attributes. (similarity 0.87)
• Sensory science quantifies perceptual attributes through structured descriptive analysis (similarity 0.87)
• Sound attributes are independent variables manipulated to elicit dependent responses. (similarity 0.83)
• Multidimensional scaling maps perceived sound dissimilarities without labeling the perceptual dimensions (similarity 0.82)
• Triadic comparison tests can derive individual perceptual structures of sound (similarity 0.82)
• Sensory scales can be calibrated for equal intensity across different attributes (similarity 0.82)
• Psychoacoustics quantifies physical sound signals into perceived auditory attributes (similarity 0.81)
• Perceptual structure analysis indirectly recovers individual feature-usage in auditory perception. (similarity 0.81)
• Multi-stimulus quality tests fail when samples vary in several perceptual dimensions at once (similarity 0.81)
• Subjective listening tests map manipulated acoustic variables to measured responses. (similarity 0.81)
• Thresholds and auditory scales link physical sound attributes to perceived auditory events (similarity 0.80)
• Perceptual attributes map onto multiple interacting physical sound properties (similarity 0.80)
• Trained listener panels trade real-world representativeness for analytic test resolution (similarity 0.80)