Thurstonian Models and Variance I: Experimental Confirmation of Cognitive Strategies for Difference Tests and Effects of Perceptual Variance

Abstract

For the triangle, duo‐trio, same‐different and 2‐AFC methods, using a model system, mean d′ values for the same subjects, discriminating between the same taste stimuli, were not significantly different. This confirmed the postulated cognitive strategies used for these methods in their respective Thurstonian/signal detection models. Introduction of perceptual variance as a result of the effects of sequences of tasting within a test, forgetting stimulus perceptions and τ‐criterion variation resulted in the 2‐AFC eliciting a significantly higher d′ than the other three methods. Yet, after a warm‐up procedure, which not only significantly increased values of d′ for all methods but also aligned subjects' τ‐criteria, the same‐different test had a d′ comparable to that of the 2‐AFC, while both d′ values were significantly higher than those of the triangle and duo‐trio. This suggested that effects of memory were more important those of sequence of tasting.

Decision strategies for the A Not-A, 2AFC and 2AFC-reminder tasks: Empirical tests

“…Lee, van Hout and Hautus (2007) presented only the reminder stimulus prior to testing, whereas both stimuli were tasted beforehand in the current study. The implication here is that tasting both stimuli may have been beneficial to the judges' performance in the subsequent task (Kim, Jeon, Kim, & O'Mahony, 2006;Lee, van Hout, & O'Mahony, 2007). This explanation has been supported by a study of familiarization effects.…”

“…This explanation has been supported by a study of familiarization effects. Kim, Jeon, Kim, and O'Mahony (2006) ascertained that experiencing the characteristics of all possible testing stimuli influences the judges' decision strategy. Sufficient familiarization with the stimuli could increase the likelihood of absolute judgment; that is, the judgment is more likely to be made by categorizing the stimuli into discrete conceptual classifications (b strategy) rather than by relying on the perceptual difference (s strategy).…”

Difference test sensitivity: Comparison of three versions of the duo–trio method requiring different memory schemes and taste sequences

“…Although SSA predicts superior performance of the triangle test over the same-different test, based on sequence effects, it has been found that triangle test resulted in equal or lower d 0 values than same-different test, suggesting the importance of memory effects (Kim, Jeon, Kim, & O'Mahony, 2006;Kuesten, 2001;Lau et al, 2004;Rousseau & O'Mahony, 2000, 2001Rousseau, Meyer, & O'Mahony, 1998;Rousseau, Rogeaux, & O'Mahony, 1999;Rousseau, Stroh, & O'Mahony, 2002;Stillman & Irwin, 1995). Yet, memory effects would not have appeared to dominate sequence effects when the duo-trio and same-different test had comparable d 0 values (Kim et al, 2006;Rousseau et al, 1998). Again, this comparison is complicated by tau-variance with the same-different test.…”

Sensory discrimination by consumers of multiple stimuli from a reference: Stimulus configuration in A-Not AR and constant-ref. duo-trio superior to triangle and unspecified tetrad?

“…This discrepancy in operational discrimination power can be attributed not only to the larger number of tests possible for the A-Not AR but also the increased cognitive load in the triangle test and the unspecified tetrad test that required a larger number of stimuli in a test than the A-Not AR method. Previous studies reported that a test method using 3 stimuli resulted in lower performance than a test method using 2 stimuli (Kim, Jeon, Kim, & O'Mahony, 2006;Lau et al, 2004). The extent of the cognitive load required for the test is an important factor affecting performance in discrimination tests ( Rousseau, Rogeaux, & O'Mahony, 1999;Rousseau et al, 2002).…”