One of the most hair-raising thought experiments in philosophy of mind is the philosophical zombie, a being that is physically and behaviorally identical to a normal human, but is not conscious of having any mental states, like the fellow on the right in the figure below.

The existence of such a being provides a strong argument against physicalism, the idea that mental states can be reduced to physical states. Many philosophers are content with entertaining the mere logical possibility of such a being and its philosophical consequences. Might such a being exist, who can differentiate red and blue like anyone else, but who has no conscious mental state corresponding to the recognition of a color?

But thinking about zombies can also be useful for psychologists interested in empirical matters, as it raises questions about the functions of consciousness. For example, what visual functions do we retain when we are unaware of seeing something? What functions do we lose? Can we see anything without awareness?

Recent research has shown that some functions once thought to require conscious processing may be performed without or with minimal awareness. For example, it may be possible to perform crude arithmetic without being aware of what one is seeing. Obviously, though, people are better at arithmetic when they are aware of seeing what it is they are adding and subtracting. This is why advertisers tend not to rely on subliminal messages (once thought to be so threatening that laws were enacted banning them). Subliminal messages may have some effect, but plainly seen messages work much better.

But might there be cases where something presented unconsciously has a larger effect than something that a person is aware of?

In a new study published in the Psychonomic Society’s journal Attention, Perception, & Psychophysics, Researchers Lin and Lu used a clever variation on a commonly used method of measuring attentional processing to show that in some cases being aware of something—namely, an attentional cue—renders it ineffective.

The basic procedure is shown below. The participants’ task was about as simple as it gets: press a button as soon as you see a dot. On baseline “no-cue” trials, participants simply pressed a button as soon as they detected the dot. On single-cue trials, the dot was preceded by a cue (the red outline shown in the figure below). For people assigned to the “informative” condition, the cue was 80% valid. That is, 80% of the time if the red outline appeared on the left, the to-be-detected dot appeared on the left. (The remaining 20% of the time the opposite happened and the cue was misleading). For people assigned to the “uninformative” condition, the single cues were non-predictive (if the cue appeared on the left, the dot appeared on the left or right with equal probability). Finally, on double-cue trials, the dot was preceded by two cues, one on each side of the display. Because the double-cue trial is necessarily spatially uninformative, it is thought that improved performance typically observed on such trials indicates that the cues act to alert people that a dot is coming up. In Experiment 1, participants were aware of all cues. We’ll get to the invisible cue condition in a moment.

The results replicated the well-known attentional cuing effect: single cues, particularly the informative ones, initially sped up dot-detection (the green triangles are below the blue triangles in the figure below). The reversal as more time passes between the cue and the dot is the signature of inhibition of return—people are slower to re-attend to a previously attended location.

The more interesting bit is shown in the bottom two panels.  The double-cues improved performance more (black triangles below white triangles) for the people in the informative group more than the uninformative group even though seeing two circles could not possibly help any of the participants to know where the dot would occur. It appears then that when single cues are, in general, uninformative, they become discounted and rendered ineffective.

A still bigger surprise came in Experiment 2 which was a partial replication of the uninformative condition of Experiment 1, but added “invisible cue” trials in which the red outline was black. This ensured that when the dot appeared, the previously presented outline cue was effectively made invisible because the dot appeared inside an identical black circle. The single cues affected people’s behavior regardless of whether they were seen. As in Experiment 1, the double-cues did not affect behavior because (it is argued) they were presented in an uninformative context and therefore inhibited. But when the double-cues were invisible they started to have a facilitatory effect! It seems that when the cue is invisible, no inhibition occurs and the attentional system just responds to it. A similar argument for a regulatory role of awareness was also recently made by Lapate and colleagues.

So there you have it. Someone—a zombie, say—can be influenced by a visual stimulus without needing to be aware of it. We already knew that. But under normal circumstances, consciously perceived cues lead to stronger effects than cues of which one is not aware. What this study shows is that the influence of an unconscious stimulus can, in some cases, be stronger than the influence of something that is consciously perceived.

Article focused on in this post:

Lin, Z., & Lu, Z. (2016). Automaticity of phasic alertness: Evidence for a three-component model of visual cueing. Attention, Perception, & Psychophysics, 78, 1948-1967. DOI: 10.3758/s13414-016-1124-5.