Within‐trial Contrast: When Is a Failure to Replicate Not a Type I Error?

Abstract

Vasconcelos, Urcuioli, and Lionello-DeNolf (2007) report the results of five experiments that fail to replicate the results of our within-trial contrast study (Clement, Feltus, Kaiser, & Zentall, 2000) and suggest that our results may represent a Type I Error. We believe that this conclusion is not warranted because (a) there is considerable evidence in support of the effect and (b) the amount of training that they gave to their pigeons prior to test may not have been sufficient to observe the effect reliably. We suggest that when sufficient training is provided, reliable contrast can be found.

Deprivation level and choice in pigeons: A test of within-trial contrast

“…Zentall and Singer (2007b), for instance, have surmised that preferences indicative of within-trial contrast can be slow to develop in pigeons (e.g., Singer et al, 2007) and, thus, may require at least 30-40 training sessions (3,000-4,000 trials at 100 trials/ session) to become apparent. However, Arantes and Grace (2008, Experiment 2) recently reported that they were unable to obtain evidence for such preferences even after more than 100 sessions (~10,000 trials).…”

“…Besides calling into question the reliability of the work ethic effect (see Vasconcelos, Urcuioli, & Lionello-DeNolf, 2007aZentall & Singer, 2007b), our inability to replicate this particular behavioral preference suggests that other proposed (or likely) manifestations of within-trial contrast should be examined. In this article, we test another prediction of this hypothesized mechanism using a manipulation we feel should also generate the required ingredients to induce behavioral preferences-deprivation level during learning.…”

Rats place greater value on rewards produced by high effort: An animal analogue of the “effort justification” effect

“…Their failure to find a significant effect of effort on reward value may have been due to the vast differences between the operant training context and the maze context used for testing, which may have prevented any effort-induced changes in flavor preference from transferring to the test context. Indeed, as noted by Zentall and Singer (2007), by the time the rats had learned about the new test context, any effort-induced flavor preference obtained during operant training may have reached extinction. In addition, Jellison (2003) noted that rats showed poor discrimination between grape-and bacon-flavored pellets in preliminary experiments.…”

The sunk cost effect in pigeons and people: A case of within-trials contrast?

“…The effect has not always been replicated (e.g., Arantes and Grace, 2008a;Vasconcelos et al, 2007a), perhaps because lower probability of reward for correct choices in training may reduce the size of the effect (Gipson et al, 2009). We do not review the debate about whether the behavioral effect can be replicated (Arantes and Grace, 2008b;Vasconcelos et al, 2007b;Vasconcelos and Urcuioli, 2008;Zentall, 2008;Zentall and Singer, 2007) because the present focus is on within-trial contrast as a theoretical mechanism that can be applied to the sunk cost effect, as opposed to a behavioral effect.…”