Recent studies suggest that performing a distraction task concurrently with a complex decision paradoxically improves the quality of the decision (e.g., Dijksterhuis et al. Reference Dijksterhuis, Bos, Nordgren and van Baaren2006b). This finding, referred to as the unconscious-thought effect (e.g., Strick et al. Reference Strick, Dijksterhuis, Bos, Sjoerdma, van Baaren and Nordgren2011) or as the deliberation-without-attention effect (see below), has been interpreted as evidence that complex decisions are best made through “unconscious thought.”

In early theoretical accounts, Dijksterhuis and colleagues defined unconscious thought as thought processes occurring outside conscious awareness and proposed that unconscious thought is capacity-unlimited (Dijksterhuis Reference Dijksterhuis2004; see also Dijksterhuis & Nordgren Reference Dijksterhuis and Nordgren2006). Thus, unconscious thought is allegedly capable of handling a large amount of information without requiring attention or cognitive resources.

We agree with Newell & Shanks (N&S) that the unconscious-thought effect, like many others purporting to demonstrate the capacity of unconscious processes, does not mandate the existence of a powerful unconscious system that works just like the conscious system, only minus consciousness (Cleeremans & Jiménez Reference Cleeremans, Jiménez, French and Cleeremans2002). We also confirm that the effect is difficult to replicate (Waroquier et al. Reference Waroquier, Marchiori, Klein and Cleeremans2009; see also Klein et al. Reference Klein, Doyen, Leys, Magalhães de Saldanha da Gama, Miller, Questienne and Cleeremans2012). However, in contrast to N&S's position, we have to admit that the results of the most recent meta-analysis (Strick et al. Reference Strick, Dijksterhuis, Bos, Sjoerdma, van Baaren and Nordgren2011) support the existence of the effect, and even more importantly, that it can be replicated under specified conditions. One of the important moderators we identified is the difficulty of the distraction task: The effect is more likely to occur with easy distraction tasks. Two independent sets of experiments that directly manipulated the difficulty of the distraction task now provide evidence that low-demanding distraction improves complex decision making compared with deliberation, high-demanding distraction, or immediate choices (Abadie et al. Reference Abadie, Waroquier and Terrier2013; McMahon et al. Reference McMahon, Sparrow, Chatman and Riddle2011). Similarly, a recent study suggests that low-demanding distraction improves creativity as well (Baird et al. Reference Baird, Smallwood, Mrazek, Kam, Franklin and Schooler2012). Abadie et al. (Reference Abadie, Waroquier and Terrier2013) also provide evidence that the effect is accompanied by a better organization of the information conveyed by decision cues, as shown by an increase in gist memory (see Reyna Reference Reyna2012).

Do such findings confirm the existence of unconscious thought, or on the contrary, do they support N&S's claim that evidence for the intervention of unconscious processes in the utilization of decision cues is lacking? As it is often the case with interesting research topics, things are not so simple.

First, we note that neither unconscious-thought theory (UTT) in its original formulation (Dijksterhuis & Nordgren Reference Dijksterhuis and Nordgren2006) nor the position taken by N&S in the target article suffice to account for these recent results. Indeed, UTT assumes that unconscious thought, unlike conscious thought, is not capacity-limited. It is precisely this greater, potentially unlimited capacity that would explain why people make better decisions after distraction. A straightforward prediction of UTT is therefore that performance should increase as distraction increases, for the more conscious attention is otherwise engaged, the more unconscious thought should be able to solve the problem. However, several studies have shown exactly the opposite: People indeed make better decisions after low-demanding distraction than after deliberation, but performance after high-demanding distraction is worse than after low-demanding distraction.

Turning now to N&S's proposal, we likewise run into an impasse, for they make the exact opposite – but likewise monotonic – prediction about the effects of distraction on decision making: The more conscious attention people can dedicate to a decision-making process, the better the decision should be. However, this is again not what the most recent studies show.

A solution to this dilemma comes from the important consideration that consciousness (i.e., the ability of being aware of the processes) and attention (i.e., the extent to which cognitive resources are allocated to process some specific information) are not identical to each other. Several studies have now documented the fact that the two can be wholly dissociated (Dehaene et al. Reference Dehaene, Changeux, Naccache, Sackur and Sergent2006; Koch & Tsuchiya Reference Koch and Tsuchiya2006; Dijksterhuis & Aarts Reference Dijksterhuis and Aarts2010). A recent experiment even adopted a clear 2 × 2 design (Watanabe et al. Reference Watanabe, Kang, Yusuke, Ueno, Asamizuya, Tanaka and Logothetis2011) to show that consciousness (as assessed by reportability) and attention are subtended by distinct neural correlates.

Based on this distinction, Strick and colleagues (Reference Strick, Dijksterhuis, Bos, Sjoerdma, van Baaren and Nordgren2011) maintained that the processes involved in the unconscious-thought effect occur outside conscious awareness but conceded that the term “deliberation without attention” (e.g., target article; Dijksterhuis et al. Reference Dijksterhuis, Bos, Nordgren and van Baaren2006b) is a misnomer. They thus suggest that the integration of decision cues is unconscious but needs attentional resources. According to this view, unconscious thought can not only perform the same functions (e.g., mathematical calculations; for evidence, see Ric & Muller Reference Ric and Muller2012) as conscious thought but also is subject to the same constraints (i.e., limited capacity).

However, Strick et al. (Reference Strick, Dijksterhuis, Bos, Sjoerdma, van Baaren and Nordgren2011) do not show any evidence that decision cues are indeed used unconsciously (e.g., cues utilization cannot be reported) during low-demanding distraction (e.g., while listening to music). Moreover, since the decision goal is given explicitly before distraction in the paradigm at hand, it is likely that some decision-relevant processes occur inside conscious awareness. If, as N&S do, we endorse the null hypothesis that complex decisions are made consciously (for evidence, see Baumeister et al. Reference Baumeister, Masicampo and Vohs2011), it seems both safe and parsimonious to assume that integration of decision cues occurs inside conscious awareness but is modulated by attentional factors. Congruently with Waroquier et al.'s (Reference Waroquier, Marchiori, Klein and Cleeremans2010) conclusions, conscious thought may enhance decision making; however, while a certain amount of attention is necessary, too much attention may sometimes be detrimental.