Thursday, 27 June 2013

How pure irrelevance turns things invisible


A faint tap per se is not an interesting sound; it may well escape being discriminated from the general rumor of the world. But when it is a signal, as that of a lover on the window-pane, it will hardly go unperceived. William James, 1890 [p. 418]

There's such a blizzard of sensory information out there, the brain would be overwhelmed if it weren't for a spotlight process of selective attention that allows us to focus. This means that once we're tuned into certain aspects of the environment, we're left blind to events outside of our selective attention - a phenomenon called "inattentional blindness".

Related to this is the idea of attention as a finite resource. It's partly because our processing powers are depleted by the focus of our attention that we're left blind to that which we ignore. A new study builds on the finite resource element of this story. Baruch Eitam and his colleagues propose that pure irrelevance is enough to render information invisible even if we have plenty of resources available for processing that information. It brings a new spin to our understanding of "induced blindness" that's not just about attentional load but also about salience and motivation.

The researchers presented 100 participants with a central coloured ring surrounded by a larger ring of a different colour. The rings appeared for half a second and were close enough together that they both fell on the high-acuity foveal region of the eye. Before the rings appeared the participants were first instructed to concentrate on either the inner or outer ring. After the rings disappeared, half a second passed and then they were asked to identify which of three distinct colours either the inner or outer ring had been. That is, they were either quizzed about the ring they'd concentrated on or the other one.

As you might expect, the participants achieved nearly perfect performance when identifying the colour of the ring they'd concentrated on (the error rate was just 3 per cent; not significantly different from flawless accuracy). In contrast, they were wrong 25 per cent of the time when trying to identify the colour of the "irrelevant" ring they'd ignored - this is almost seven times the error rate compared with the "relevant" ring. Participants were also less confident in their knowledge about the irrelevant ring.

A follow-up study confirmed that blindness to the ignored ring colour was induced purely by its irrelevance, not because of limitations in attentional resources. Twenty-six participants performed an identical task to before except this time they were instructed to concentrate on both coloured rings. Regardless of which ring they were subsequently quizzed on, their performance was virtually flawless (the error rate was 8 per cent: not significantly different from performance for the relevant ring in the first experiment).

Eitam and his colleagues said their research suggests there are two kinds of induced blindness - the first involves the unavailability of processing resources and is accompanied by a lack of visual awareness, so the missed information is literally not seen. The second, demonstrated here, is based on irrelevance alone and is related to not having sufficiently processed information that was seen.

"We propose that in our study the participants who failed to report the irrelevant colour were aware of it when it was displayed," the researchers explained, "but due to irrelevance, its abstract representation (of 'red' for example) was insufficiently activated, or possibly suppressed, and the participants could not report it." This an interesting extension of established psychological theory that will surely be of interest to road safety researchers, advertisers and magicians.

_________________________________ ResearchBlogging.org

Eitam, B., Yeshurun, Y., and Hassan, K. (2013). Blinded by irrelevance: Pure irrelevance induced “blindness”. Journal of Experimental Psychology: Human Perception and Performance, 39 (3), 611-615 DOI: 10.1037/a0032269

Post written by Christian Jarrett (@psych_writer) for the BPS Research Digest.

7 comments:

Anonymous said...

A good explanation of why women say that they become invisible to younger men once they hit 40. Young men simply find them irrelevant.

Daniel Simons said...

Christian - this seems like an odd framing for this finding. I don't know the actual result or if they made this argument in their paper, but as you describe it, it does not challenge the existing literature on inattentional blindness, and it isn't really even that novel in that context either.

As far as I know, nobody in the inattentional blindness literature claims that inattentional blindness only occurs when attention is fully occupied. A number of studies show that noticing rates drop when the primary task is harder, but plenty of people miss the unexpected object even when the primary task is easy. The point is just that people are focused on something, not that their task uses up all available resources.

Moreover, noticing rates vary substantially depending on the relationship between the unexpected object and the unexpected object shows that it's not just a matter of resource limitations (e.g., people are more likely to notice a black circle when attending to circles than when attending to squares, even when the attention task is equally demanding). If the failure to notice were entirely resource based, then it shouldn't matter which shapes were attended. Evidence from studies of individual differences shows that differences in the ability to track the moving objects does not predict noticing rates for unexpected objects. That means that even if people find the tracking task easier (i.e., they have more resources available), they are no more likely to notice the unexpected object. In all of these cases, the unexpected object is irrelevant as well, so the irrelevance aspect of this finding isn't entirely novel either.

It's also not clear to me from the description whether this is a study of inattentional blindness. Inattentional blindness is a failure to notice an entirely *unexpected* object or event when attention is focused on something else. if people knew that the outer ring was going to be there (or if they did this on multiple trials), then it is not a study of inattentional blindness. Rather, it is a study of how well people filter content they know to be irrelevant. It isn't a second type of inattentional blindness — it's a demonstration of the power of focused attention to filter details known to be irrelevant. That's important, of course, but it doesn't necessarily tell us anything about inattentional blindness.

Christian Jarrett said...

Hi Dan, thanks for providing this extra information. I hope I did present the position of Baruch Eitam and his colleagues correctly.

They write: "prior demonstrations of induced blindness always involved situations in which resources were depleted" (e.g. Simons & Chabris, 1999). They add: "these demonstrations were designed to ensure that the main task the participants had to perform (i.e., the explicitly instructed task) was demanding enough, perceptually and/or cognitively, to ensure that available resources are engaged with the processing of task-relevant information. Thus far then, induced blindness is generally accepted to be due to the ‘draining’ of resources by relevant stimuli. Yet this leaves open the possibility that irrelevance alone could be sufficient for creating blindness."

I will email you the paper so you can examine in more detail if you want to.

Daniel Simons said...

Thanks for emailing the paper. From a quick read, I'm afraid that there are a couple of important mischaracterizations in your post.

They aren't claiming that there are two types of inattentional blindness. Rather, they are claiming that there are different forms of *induced* blindness. ("we suggest that there might be two types of induced blindness. Both depend on relevance but only one depends on relevance alone.") That's a critical difference, and one they make clear in discussing their findings. Inattentional blindness is a particular type of "induced blindness," but there are others (e.g., change blindness).

They are showing that you can have induced blindness for an irrelevant stimulus. They aren't claiming that people fail to see the irrelevant ring entirely. They are saying that people don't take notice of its features or remember them. Inattentional blindness is a failure to notice an unexpected object or event. They aren't studying inattentional blindness here.

Although they do argue against a "strong selection" idea, it's something of a straw man in the context of IB, and they seem to know that. It's not central to their claims or finding. Basically, they are arguing that, *if* you hold a strong selection account for IB (which pretty much nobody does), then their finding suggests another possible explanation for inattentional blindness. Their finding shows that even when attention is not particularly taxed, we don't tend to encode/remember the details of irrelevant object.

Their finding is interesting, but it really is a study of how we don't remember things that are outside the focus of attention and known to be irrelevant rather than a study of inattentional blindness per se. They don't describe it as an inattentional blindness study in the paper itself.

Christian Jarrett said...

Understood, thanks Dan - I made some tweaks.

baruce said...

Hi Christian & Dan,
Christian, many thanks for the interest and nice depiction of our work. Dan, my only comment on your narrative is that we do not see the results as stemming from knowing something and then forgetting it. Rather, on the basis of our previous theoretical work, we hypothesized that relevance would be necessary for activating semantic knowledge related to a stimulus. For example, if people saw the color red in an irrelevant location, the category 'red' would not be activated in their mind (or less so) and in a real sense would not know what they saw. Which is what we indeed found.

In another paper just accepted at the same journal we show that no trace of the irrelevant knowledge is left even after multiple exposures.

In the 'irrelevance induced blindness' paper we actually adopted a descriptive term coined by Dan in a Current Directions paper from 2000 -- 'Inattentional Agnosia' -- although I would personally opt not to use the term 'attention'. But that's another (long) story.

Cheers,
Baruch Eitam

Christian Jarrett said...

hi Baruch, thanks for your comments and the extra background info.

Post a Comment

Note: only a member of this blog may post a comment.

Google+