The Effect of Background Feature on Time to Locate Target Letters
Grace Riley, Sukey Steckel, Courtney Prosser, Tara Linman, and Pete Staub
Elon College
May 1998
Abstract
The ability to locate a target letter in different backgrounds was investigated in a 2 X 2 within
subjects factorial design. Twenty subjects tried to locate two different target letters, Z and Q,
in a series of angular and round backgrounds. Results indicated a higher overall reaction time
for the letter Q than the letter Z. Resulting from a significant interaction, reaction time for Q
in a round background was greater than in an angular background, and reaction time for Z in an
angular background was greater than in a round background. No main effect for background was
found.
Introduction
The human brain has the ability to scan its environment to focus on a single object. In
order to investigate scanning time, we conducted a study that focused on the way stimulus
features influence visual search/feature detection. When we scan our environment in order to
locate a specific object, we are engaging in a process referred to as visual search. One
characteristic of visual search is that irrelevant aspects of an object need not be identified fully
to search successfully.
There are various classic studies that have addressed visual search, including extensive
research by Ulric Neisser. Neisser (1954) reported a study in feature detection which
compared trials that have a set target letter as opposed to trials that do not have a set target
letter. Neisser found that scanning time was faster for those trials which had a set target letter
to search for. According to Neisser, being told to find a specific target is easier because of
template matching. In using this method, the brain creates a visual image of the target letter
and attempts to match it during the scanning process. Additionally, Neisser, Novick, and Lazar
(1963) examined the effect of the number of targets to locate on scanning time. The results
showed that subjects do not scan any faster for one target than for ten targets.
Neisser (1963) also studied the influence of specific target letters, Q and Z, and the
amount of letters in each display (2 vs. 6) ,on the time it took for subjects to locate the target
among different backgrounds. The independent variables were the type of target letter, the
number of letters in each display (2 vs. 6), and the complexity of processing required. The
target letters being looked for, were Z and Q, and were arranged in the angular backgrounds
(EIMVWX) and round backgrounds (CDGORU). The results of this study showed that contextual
background is very imposing on processing time of angular and round letters. In the angular
context, it was found that location of the letter Q took much less time than the letter Z. The
opposite was found for the round context.
Our study consisted of two purposes. The first purpose was to investigate the impact of
various types of background features on the time it takes to locate target letters which are
similar or dissimilar to the backgrounds. Our second purpose was to explore the differences in
the search times for the angular and rounded targets, regardless of background features. In
doing so, we had subjects participate in a computer task which required them to locate a specific
target among a display of various letters.
We predicted that there would be a higher search time for locating angular target letters
located in angular backgrounds, and that there would be a higher search time for locating round
target letters among rounded backgrounds. We also predicted that search time would be highest
when searching for a round target as compared to an angular target, regardless of background
features.
Method
Participants
Twenty participants were chosen for participation on a voluntary basis from various
undergraduate courses. Physical restrictions included correctable 20/20 vision and the ability
to manually operate a personal computer with a keyboard attachment. Another restriction was
that a participant be able to read and understand the English language. No additional restrictions
were necessary for participation.
Apparatus
The apparatus consisted of four personal computer workstations located in four corners
of a well-illuminated laboratory room which measured approximately 40 by 25 feet. The work
stations included a chair, as well as a Quantex computer, a 13 inch MAG monitor, and a PC
compatible keyboard placed on a desk. A computer program was used to present stimulus
displays and record search (reaction) times.
Procedure
The participants engaged in a visual search feature detection task for a single session
lasting approximately fifteen minutes. At the beginning of the session, the experimenter read
aloud the task instructions to the subjects as they read them on their screens. Following the
instructions, each of the participants completed four practice trials which provided a model of
upcoming trials.
For the experimental task, participants were presented with 160 trials. Each trial
consisted of participants visually scanning rows of letters that appeared on the computer screen
moving from left to right and top to bottom. On an equal proportion of the trials, participants
scanned for an angular target (Z) in an angular background, an angular target (Z) in a rounded
background, a round target (Q) in a rounded background, and a round target (Q) in an angular
background. At the end of each trial, participants were given the opportunity to redo that
specific trial if they did not follow the instructions properly. That is, if the participant went
through a trial and located the target without reading through all prior display lines, they were
to repeat that trial. The dependent variable was the scanning time it takes to locate the targets.
Experimental Design
The design was a 2 X 2 within subjects factorial design. The first factor was the
target letter (Z or Q) and the second factor was the background feature (angular or round). A
2-factor within subjects ANOVA was used for data analysis.
Results
As shown in Table 1, there was a statistically significant main effect for letter type,
F(1, 19) = 7.50, p < .05, such that reaction time was significantly faster for Z as compared to
Q. Also, there was a statistically significant letter type X background interaction, F (1,19) =
25.59, p < .01.
Table 1
Mean Reaction Time as a Function of Target Letter and Background Features
__________________________________________________________________
Background Features
_______________________________
Round Angular
Letter
__________________________________________________________________
Q
M 5565.85 4802.15
SD 2016.26 2117.47
Z
M 4646.95 4977.05
SD 2076.99 2502.45
__________________________________________________________________
As shown, reaction time for the letter Q was faster in the angular background as compared to the
round background, however, reaction time for the letter Z was slower in the angular background
as compared to the round background. There was no statistically significant main effect for
background, F (1,19) = 2.05, p > .05.
Discussion
Our results revealed visual search performance was significantly impacted by both
target letter and background features. We found higher overall reaction time, in searching, for
the letter Q as opposed to Z. This finding is supportive of the research of
Neisser that, and is likely due to the shapes of the target letters. That is, angular letters have
more distinct features, and the greater number of distinctive features may lead to greater
ability to locate the letter regardless of background. Also, in support of our interaction
prediction, we found that the letter Z was more quickly located in a round background as
opposed to an angular background, and the letter Q was more quickly found in an angular
background as opposed to a round background.
This pattern of quicker location of target letters in a dissimilar background is consistent
with the research of Ulric Neisser (Neisser, 1963, 1964). By putting an angular target in a
rounded background or rounded target in an angular background, scanning time decreases. This
finding suggests that placing a target in a dissimilar background serves to facilitate individuals'
perceptions of the target's individual features as it "pops out" of the background.
Future research may extend these findings in an attempt to understand the impact that
instructions may have on the search process by comparing performance of those instructed to
search in a row by row manner as our subjects were instructed, to the performance of
individuals instructed to search in a method emphasizing searching as quickly as possible.
Additionally, future research may address the impact of gender on scanning time or instruction
method.
In an applied sense, the results of our study and similar research could be useful for
those in the world of advertising, public relations, and safety and traffic signs. They could use
our findings to help determine the most beneficial ways to advertise products in order to catch
the consumer's eye or to alert people of safety precautions.
References
Neisser, U. (1963). Decision time without reaction time: Experiments in visual
scanning. American Journal of Psychology, 36 , 376-385.
Neisser, U. (1964). Visual search. Scientific American, 17, 94-100.
Neisser, U., Novick, R., & Lazar, R. (1963). Searching for ten targets
simultaneously. Perceptual and Motor Skills, 17, 955-961.
Return to Top of Page
Return to Undergraduate Psychology Papers Contents