Source: Laboratory of Jonathan Flombaum—Johns Hopkins University
We generally think that we see things pretty well if they are close by and right in front of us. But do we? We know that visual attention is a property of the human brain that controls what parts of the visual world we process, and how effectively. Limited attention means that we can't process everything at once, it turns out, even things that might be right in front of us.
In the 1960s, the renowned cognitive psychologist Ulrich Neisser began to demonstrate experimentally that people can be blind to objects that are right in front of them, literally, if attention is otherwise distracted. In the 1980s and 1990s, Arien Mack and Irvin Rock followed up on Neisser's work, developing a simple paradigm for examining how, when, and why distracted attention can make people fail to see the whole object. Their experiments, and Neisser's, did not involve people with brain damage, disease, or anything of the sort, just regular people who failed to see objects that were right in front of them. This phenomenon has been called inattentional blindness. This video will demonstrate basic procedures for investigating inattentional blindness using the methods of Mack and Rock.1
1. Stimuli and design
- The stimuli for this experiment can be made with basic slide software such as PowerPoint or Keynote.
- The first stimulus to make is called the noncritical stimulus.
- On a white slide, create two black lines; the first should be about 80% of the vertical length of the whole slide, and the other just slightly longer. In Figure 1a, the slide is 770 px long, the shorter line is 630 px long, and the other is 645 px.
- Now choose the shorter of the two lines and rotate it by 90° so that it is horizontal, and center the two lines in the middle of the screen so that they form a cross, as in Figure 1b.
- Duplicate the slide just made, but make the shorter line the vertical arm of the cross, and the longer line the horizontal arm of the cross, as in Figure 2.
Figure 1. (a) Two lines that are used to construct the cross stimulus in (b). The line on the left is slightly shorter than the one on the right, a difference that is easy to see when they are aligned and oriented vertically, but difficult to see when they are oriented to form a cross. The cross in (b) is an example of the noncritical stimulus. The task of participants is to judge which line of the cross is longer. (In the case shown, the vertical line is longer). The difficulty of this task draws on attention.
Figure 2. An example of a noncritical stimulus. In this example, the horizontal line is the longer one. Seeing the difference should be very difficult.
- Now make the critical stimuli.
- Just duplicate the two noncritical stimuli. In the first, create a small grey star and place it in one of the four quadrants, in Figure 3a the star is in the lower-right. In the second critical stimulus, make a small grey triangle, and place it in any of the quadrants. In Figure 3b the triangle is in the upper left quadrant.
Figure 3. Two examples of critical stimuli. Each of the stimuli has a shape in one of the quadrants defined by the cross. The question of interest in the experiment will be whether observers see this shape under various conditions of attentional engagement with the task.
- Finally, make the fixation stimulus and the mask.
- The fixation stimulus is just a blank slide with a small cross in the center, 20 x 20 px.
- To make a mask, set each pixel randomly to black or white. Figure 4 shows an example.
Figure 4. A mask stimulus. In the mask, each pixel or square in the slide is set randomly to black or white. The purpose of a mask like this is to flush previous stimuli from the visual system. It allows experimenters to finely control the amount of time that an observer is exposed to a specific stimulus. This is because activity in retinal cells and brain cells can persist, even after a stimulus is absent. A blank screen-especially a dark one-lets activity persist for an especially long time, even producing afterimages. A mask, like the one shown, randomly rearranges all the firing in visually responsive neurons rather than allowing their prior activity to persist after the stimulus has been removed.
- All that remains is to assemble the images just created into trials. There are two types of trials. To make a noncritical trial, present the fixation stimulus for 1500 ms, followed immediately by one of the noncritical stimuli for 200 ms, followed immediately by the mask for 500 ms. Figure 5a schematizes the sequence.
- The critical trials are identical to the non-critical ones with one exception: Include, here, a critical stimulus for 200 ms, instead of the noncritical stimulus. Figure 5b schematizes the sequence of events.
Figure 5. Schematic depictions of the sequences of events in (a) noncritical and (b) critical trials. The only difference between the two trial types is which stimulus is shown in the middle for 200 ms, the critical or the noncritical. Each block of the experiment will include three trials, two noncritical trials followed by a critical one.
- Experimental design: The experiment will ultimately involve a total of nine trials, in three groups. The first group of three trials is called the inattention set, the second group is called the divided attention set and the third is called the complete attention set.
- Each group of trials will involve two noncritical trials followed by a critical trial. The only difference between the three groups of trials is in the instructions given to participants.
2. Running the experiment
- An experiment requires at least 50 participants, but each session is very short. The original experiments were run by recruiting participants at a science museum. A library or a campus quad on a nice day are also good places.
- To test a participant, use the following procedure:
- Ask someone if they are willing to participate in a very short experiment on visual perception.
- When they say yes, point to the computer screen, with the fixation stimulus already present, and say the following: "I'd like you to just point your eyes at this cross, without moving them. In a moment, the cross will be replaced by a larger cross, but in that one, one of the lines will be just a little longer than the other. I'd like you to carefully look at that cross, and try to report which one is longer, the horizontal or the vertical line. The cross will be present very briefly-for much less than one second-so you really need to look closely when you have the chance."
- Ask the participant if she has questions, and after answering any, run the first noncritical stimulus of the inattention set. Write down whether the participant reported the horizontal or vertical line as the longer.
- Now run the second noncritical trial, and then run the critical trial to complete the inattention set.
- After the critical trial, ask the following question, and record the participant's answer: "In that last trial, or any of the others, did you see anything else on the screen besides for the test cross?" If the participant says, 'yes', ask her to describe what they saw and where she saw it, recording the complete answer.
- Now it is time to run the divided attention trials. To begin, just say the following to the participant: 'I'd like you now to do three more trials. Are you ready?"
- Again, run two noncritical trials, followed by a critical one. After, the critical trial, ask the participant if they saw anything unexpected, what it was, and where.
- Finally, run the complete attention trials, by saying the following: "I'd like you to do three more trials. But this time, you don't need to tell me which of the two lines is longer. Instead after each trial, just tell me whether you saw anything besides for the two lines, what it was, and where on the screen it was.
- Run two noncritical trials followed by a critical one.
3. Data analysis
- To analyze the results, look at the responses given by each participant to each of the critical trials. Remember, those are the trials that have a shape somewhere in the display along with the large cross. Mark whether the participant saw the stimulus, counting it as seen if the participant reported the shape or the quadrant that it appeared in correctly.
- Sum up the number of participants who saw the stimulus in each set of trials (inattention, divided attention, and complete attention).
We don’t always process the entirety of our physical surroundings—especially when our attention is too focused—which can affect what we perceive and ultimately see.
In a given environment, a person can be simultaneously exposed to different visual stimuli—including posters on a wall, components of a gaming system, or virtual zombies on a TV screen.
If one of these items is related to a challenging perceptual task—such as targeting multiple, advancing undead to beat a high score—an individual will focus on it.
As a result, limited or no attention is paid to preexisting objects in the room, or any novel thing that enters it—like a significant other who walks in.
Such lack of attention means that the game player’s brain does not effectively process the visual stimulus of their partner, and thus they do not see them. This phenomenon of having a salient object in view, without attending to it and therefore not seeing it, is called inattentional blindness.
Using the techniques of Arien Mack and Irvin Rock, this video explains how to generate stimuli, collect and interpret data, and it notes how researchers are studying inattentional blindness today.
In this experiment, participants are exposed to three trial conditions of attentional engagement—inattention, divided, and complete—and, within each, asked to report what they see.
Independent of condition, a single trial consists of three sequential components: a fixation point, the stimulus, and a mask. The first element, the fixation symbol, consists of a small, centrally-positioned cross that serves as a focus point for participants’ eyes.
This is followed by the stimulus, which can be either non-critical or critical. Although both consist of a large, centered test cross—much bigger than that shown previously—the critical stimulus contains an additional gray shape in one of the quadrants.
The trick here is that the two perpendicular lines, regardless of the type of stimulus, are different sizes: one is slightly shorter than the other.
Importantly, the longer of these two marks must be identified—a difficult objective that requires visual attention.
Since the gray shapes will be onscreen during critical trials, these items are meant to assess inattentional blindness—whether a participant reports seeing them.
The final component of a trial, the mask, consists of a grid in which squares are randomly set to black or white. This mottled image serves to flush the previously shown stimulus from the visual system.
The first condition of the experimental task, inattention, involves participants being shown the three components of a non-critical trial on a computer monitor, after which they must state which of the two perpendicular lines in the test cross is longer.
Afterwards, a second non-critical trial followed by a critical one are presented. The idea is that, since the goal is to identify the longer lines in the stimuli, participants dedicate the majority of their attention to the test crosses onscreen. As a result, limited attention is paid to the gray shape shown in the third, critical trial—it is being inattended to.
When the three trials have been completed, participants are asked whether—in any of the test crosses shown—they saw an unexpected object.
Here, and in subsequent conditions, the dependent variable is the number of participants that accurately state either the type of shape shown during the third trial—in this instance, a star—or the quadrant it falls in.
Based on previous research, it is expected that the majority of participants will report that they did not see any objects—aside from the lines in the crosses—during the trials, providing evidence for inattentional blindness.
The next condition, divided attention, follows the same format: longer lines must again be identified in two non-critical and then a subsequent critical trial.
However, the trick is that, since participants were questioned about unusual objects at the end of the inattention condition, they’ll now be on alert for such out-of-place items. In other words, their attention will be divided between identifying larger lines in crosses, and looking for odd images.
It is anticipated that—after the three trials in this group are presented—more participants will indicate that they saw a new gray shape compared to the inattention condition, emphasizing the role that attention plays in visual perception.
The final condition is complete attention, and—in contrast to previous sets—it is stressed that the long lines do not need to be distinguished. Rather, the only goal is to name any objects that appear onscreen during the trials, along with their location amongst the quadrant.
Other than the directions presented at its start, the format of this group is the same, and again involves two non-critical trials followed by a critical one.
As they are told to only focus on items other than the perpendicular lines, it is expected that participants’ complete attention will be concentrated on the gray shape that appears in the critical stimulus, and—similar to the divided attention condition—the majority of them will indicate that they saw it.
To prepare the stimuli for the experiment, begin by opening basic slide software on a computer. On a white background, proceed to draw a single, vertical line that is approximately 80% of the height of the slide.
Then, on the same sheet, create a second vertical line that is slightly shorter than the first—here, the smaller bar is 630 px, and the larger one 645—and rotate it 90°. Afterwards, center the two marks so that they intersect and form a cross in the middle of the screen.
Proceed to generate a second slide in the same manner, but instead rotate the longer line so that it forms the horizontal axis of the test cross. Once completed, these two slides will compose the non-critical stimuli.
To make the critical images, duplicate the sheets, and in the one containing the short horizontal line, use the shape tool to include a gray star in a random quadrant of the cross. Repeat this process for the slide with the long horizontal mark, inserting a gray triangle and a square for the third condition.
Then, on a new blank slide draw two short lines, each approximately 20 px in size. Next, arrange the bars so that they form a small cross in the center. This image will function as the fixation point.
Finally, open an additional white sheet and create the mask screen. To do this, construct a grid of repeating squares and randomly fill in some of them with black to make a checkerboard.
With all of the types of stimuli generated, arrange the order such that the first two sets of three slides in each group of three are the non-critical trials consisting of the fixation symbol, the test cross only—make sure to note where the longer lines were placed—and the mask.
For the third set in each cluster, repeat the same order, with the only difference being the contents on the second slide in the series. This should now contain the critical stimuli—both the lines and one shape.
Prior to starting the task, welcome the recruited participant and verify that they would like to take part in a short experiment on visual perception. Then, proceed to direct them to a computer screen on which the small fixation cross is already displayed.
Continue by pointing to the symbol onscreen, and instruct the participant to look at it and not move their eyes. Stress that the next slide—also with a cross—will be shown only briefly, and should be carefully studied to identify which of the two lines displayed on it is longer.
Upon ensuring that all questions have been answered, press the spacebar to initiate three trials of the inattention condition. For each, show the fixation symbol for 1500 ms, the non-critical or critical stimulus for 200 ms, and the mask for 500 ms.
Afterwards, inquire whether the participant saw additional images in any of the test cross slides, and expect that—for this condition—they will answer "No."
Record this response, and then run the three trials of the divided attention condition. Once all slides have been shown, again inquire whether the participant observed any unusual items, and anticipate that they will reply "Yes."
If they do, have the participant elaborate on what shape they observed, in which trial it appeared, and in what quadrant of the screen it was located.
After recording the divided attention data, inform the participant that they will be shown a final set of stimuli. However, stress that in this last group, they only need to report whether they see shapes aside from the crosses—the lengths of lines are unimportant.
End the experiment by running the complete attention trials, and noting what gray shapes the participant saw.
To analyze the data, for each of the three conditions—inattention, divided attention, and complete attention—calculate the percentage of participants that reported observing a gray item in the critical trial.
Keep in mind that in order for this object to be counted as "seen," the participant must either have accurately reported the shape—either a star, triangle, or square—or the quadrant in which it occurred.
Notice that, for the inattention group, only 40% of individuals reported being aware of the extra item, while the remaining 60% did not, providing evidence for inattentional blindness. Importantly, these results suggest that an item must be attended to, in order to be seen.
In contrast, approximately 95% of individuals in the divided attention group and 100% in the complete attention set observed the shapes, likely due to the fact that some of the participants’ attention was allocated to finding these items, thus enabling their brains to effectively process them.
Now that you know how visual, line-based stimuli can be employed to study inattentional blindness and what a person sees, let’s take a look at how researchers are investigating this phenomenon in other ways.
Up until now, we’ve focused on how visual-based tasks—such as judging lengths, like for cracks in a sidewalk—affect a person’s awareness of their surroundings.
However, other researchers are looking at whether talking on a cell phone—an auditory task that requires a great deal of a person’s attention—can influence what they visually perceive.
Such work has shown that pedestrians on cell phones demonstrate riskier behavior—like narrowly bumping into someone—than their non-talking counterparts.
Furthermore, these individuals even report that they fail to see outlandish stimuli that a researcher introduces into their environment—such as a clown on a unicycle—providing evidence for inattentional blindness, possibly caused by the perceptual demands of their conversation.
Other researchers are partnering with magicians—who routinely manipulate their audience’s attention during an act—to better understand different aspects of inattentional blindness.
For example, some work has paired a "disappearing" trick—whereby a performer makes an object, like a lighter, vanish into thin air—with eye-tracking technology.
When the eye fixation points of participants who claimed to have seen the lighter fall were compared to those from subjects who did not detect this action, in both instances it was found that individuals tended to focus on the magician’s face or the hand supposedly holding the flame.
These results demonstrate that it is where attention is directed—not necessarily where the eyes are positioned—that influences what a person sees.
You’ve just watched JoVE’s video on inattentional blindness. By now, you should know how different-sized lines—with or without gray shapes—can be used to assess a person’s awareness of their visual world. You should also understand how to collect and interpret visual perception data, and realize how directed attention—rather than eye position—leads to inattentional blindness.
Thanks for watching!
Figure 6 graphs the percent of participants who saw the critical stimulus in the critical trial of each of the three types of trial sets. Note that far fewer saw it in the inattention set, and more importantly, in that set only about 40% saw the stimulus at all. That means that 60 out of every 100 participants failed to see a large object right in front of them. This failure is what is called inattentional blindness. The length judgment task is difficult and uses up all of the observer's attention. As a result, there is no attention left to process the unexpected shape, and this demonstrates that seeing something requires attending to it.
Figure 6. Results of an inattentional blindness experiment including 50 participants. The primary dependent variable of interest is the percent of participants who accurately reported the position or shape of the critical stimulus in a critical trial. There was one critical trial in each set of three trials, and there were three sets: the inattention set, the divided attention set, and the complete attention set. More than half of participants failed to see the shape in the inattention critical trial, a result that demonstrates the presence of inattentional blindness.
In contrast, in the divided attention and complete attention trials the observer has already been asked about unexpected objects, or even told to look for them. As a result, the observer allocates some attention throughout the display, and this allows her to process and see the shapes presented in the third (critical) trial of each set. As the figure shows, all or nearly all participants should see the shape in the divided and complete attention critical trials.
Note that the divided attention trials receive their name because of the fact that once the observer has been asked about unexpected objects, those objects stop being entirely unexpected. It is therefore assumed that the observer will allow some attention to search the displays on those trials. The complete attention trials are named accordingly because the instructions in those trials direct the observer to focus entirely on seeing any object besides for the cross.
Applications and Summary
An important set of applications for inattentional blindness research is in the domain of driving safety. When people have car accidents, it is not uncommon for them to report that they failed to see the car, or person, or object that they hit. It makes sense to think they failed to see it because they were perhaps looking away. Inattentional blindness suggests that they could fail to see even while looking in the right place, that is, if attention is distracted. Researchers have used driving simulators, therefore, to conduct experiments on whether inattentional blindness may cause car accidents and how to reduce accidents. For example, talking on a cell phone appears to engage attention and increase the likelihood of an accident induced by inattentional blindness.
- Rock, I., Linnet, C. M., Grant, P.I., and Mack, A. (1992). Perception without Attention: Results of a new method. Cognitive Psychology 24 (4): 502-534.