Electrophysiological Evidence for the Failure of Salient Stimuli to Capture
Attention, even when Presented Rarely
Birken Noesen & Mei-Ching Lien
Eric Ruthruff
Oregon State University
N2pc Effect = ERP contralateral to cue– ERP Ipsilateral to cue
Experiment 1 (N = 16)
Purpose: Determine whether an irrelevant color singleton has the power to
capture spatial attention, producing a cuing validity effect and an N2pc, when
it appears rarely.
Target Display: Each participant responded to either the red letter, the green
letter, the blue letter, or the yellow letter (4 groups). The target displays
contained two T’s and two L’s. For the red and green target groups, one letter
was red, one was green, and two were white. For the blue and yellow target
groups, one letter was blue, one was yellow, and two were white. The two
colored letters always appeared on opposite sides of the display.
1200-1400 ms
50 ms
100 ms
T
L
L
T
50 ms
Until Response
100 ms TIME
EEG Data Analyses: EEG epochs were time-locked to singleton cue onset.
The cue-elicited N2pc was assessed 170-270 ms after cue onset; the targetelicited N2pc was assessed 350-450 ms after cue onset.
Sites: O1, O2, PO5, PO6, P5, and P6
Predictions: If a salient-but-irrelevant singleton can capture attention when
presented rarely, the cue should produce cue validity effects and N2pc effects.
Cue displays for Red/Green targets
For the blue and yellow target groups, the color singleton cue was a red box
(among green boxes) or a green box (among red boxes).
Relevant Cue Only
Behavioral Data:
Cue displays for Blue/Yellow targets
10% Different Hemifields
Relevant + Onset Cue
Results and Discussion
650
Mean RT was not significantly
Behavioral Data:
different between the no-cue
condition and the valid and invalid
650
cue conditions, Fs(1,12)≤1.78,
53
ms*
600
ps≥.2072, indicating that the salientbut-irrelevant cue did not capture
550
attention.
600
550
10% Same Hemifield
80%
527 ms
533 ms
533 ms
500
450
400
Valid
Invalid
Color Singleton Cue
No Cue
N2pc Effects (cue present trials):
* p <.0001
68 ms*
500
450
Relevant Cue
Cue-Elicited N2pc Effects
-200
N2pc Effects:
0
200
400
600
Cue-Elicited N2pc Effects
Relevant Cue
-200
3
200
400
600
800
Post-Cue Interval (ms)
-3
Cue and target in the same hemifield
Cue and target in different hemifields
-200
Relevant + Onset Cue
This research was supported by funding from Oregon State University
Undergraduate Research, Innovation, Scholarship, and Creativity to Birken
Noesen.
0
0
Post-Cue Interval (ms)
Folk, C. L., Remington, R. W., & Johnston, J. C. (1992). Involuntary covert orienting is
contingent on attentional control settings. Journal of Experimental Psychology: Human
Perception and Performance, 18, 1030-1044.
Horstmann, G., & Ansorge, U. (2006). Attentional shifts to rare singletons. Visual Cognition, 14,
295-325.
Lien, M.-C., Ruthruff, E., Goodin, Z., & Remington, R. W. (2008). Contingent attentional capture
by top-down control settings: Converging evidence from event-related potentials. Journal
of Experimental Psychology: Human Perception and Performance, 34, 509-530.
Target-Elicited N2pc Effects
-3
800
0
3
Relevant + Onset Cue
Relevant cues (in the target color)
produced large cue validity effects,
regardless of whether there was a
simultaneous abrupt onset,
ts(9)≥7.68, ps<.0001. Instead of
reducing capture by the relevant
stimulus, the onset may have actually
increased capture, F(1,8)=3.50,
p=.09.
Target-Elicited N2pc Effects
-3
Pooled
Valid
Invalid
400
Acknowledgements
OR
Design: Half of the participants responded to the red letter in the target display
and the other half responded to the green letter. The cue display contained a
box colored the same as the target letter, which would normally capture
attention. On 80% of the trials, this relevant cue appeared by itself, and on the
other 20% of trials, it had to compete with a simultaneous abrupt onset (the
relevant+onset cue). The onset appeared in the same hemifield as the relevant
cue (but not the same location) for 10% of the trials, but in the other hemifield
for the other 10% of the trials (see below for examples for the target color red).
Results and Discussion
Cue Display: 80% of the trials (1024 trials in total) contained no cue, whereas The irrelevant color singleton cue produced no N2pc effect during the period
170-270 ms after cue onset, ts(15)<1.0, suggesting that it did not capture
20% of the trials (256 trials in total) contained an irrelevant color singleton.
attention, despite being presented rarely.
For the red and green target groups, the color singleton cue was a blue box
(among yellow boxes) or a yellow box (among blue boxes).
References
OR
Purpose: Examine the ability of rare abrupt onsets (20% of trials) to capture
attention away from a relevant cue.
uV
To assess capture, we measured cue validity effects (e.g., shorter RT for valid
trials than invalid trials). We also assessed capture using an
electrophysiological measure - the N2pc - believed to reflect the allocation of
spatial attention. When attention is directed to the left or right visual field,
brain potentials are found to be more negative in the contralateral hemisphere
than in the ipsilateral hemisphere. The N2pc is the difference in event-related
brain potentials (ERPs) between these hemispheres. It is usually strongest
over posterior scalp, about 170-270 ms after stimulus onset. The N2pc effect
provides an index of whether and when attention has been directed to the cue.
100 ms
Target
uV
The goal of the present study was to determine whether salient-but-irrelevant
objects can trigger the involuntary capture when presented rarely. We used a
cuing paradigm, with non-informative cues (25% valid vs. 75% invalid).
1200 ms
Cue
Tone
Feedback
Experiment 2 (N=10)
Response Time (in ms)
The Present Study
Fixation
uV
Can salient objects capture our attention? Some have argued that the salient
stimuli, such as color singletons, do not have inherent power to capture our
spatial attention unless they contain some properties match those we are
currently looking for (contingent capture; e.g., Folk, Remington, & Johnston,
1992; Lien, Ruthruff, Goodin, & Remington, 2008). Others have argued that
frequent present presentation of the salient objects diminished their power to
capture attention (e.g., Horstmann & Ansorge, 2006). The present study
examined capture by salient objects when presented only occasionally, using
both behavioral measures (response time [RT] and proportion of error [PE])
and electrophysiological measures.
Sample Event Sequence: The target was red in this example.
Response Time (in ms)
Introduction
University of New Mexico
0
200
400
600
800
0
3
Post-Cue Interval (ms)
Cue and target in the same hemifield
Cue and target in different hemifields
The relevant cue by itself elicited a substantial N2pc effect (-0.398 µV), ts(9)≥2.19, ps≤.05. When competing with an abrupt onset, the relevant cue still
elicited a strong N2pc effect (-0.386 µV; ts(9)≥-1.33, ps ≤.22); the difference
was not significant, F<1.0.
Conclusions
Using behavioral and electrophysiological measures, we found no evidence
that salient-but-irrelevant objects (color singletons and abrupt onsets) have the
power to capture spatial attention, even when presented rarely (20% of trials).
We found that capture was driven primarily by top-down control settings, not
bottom-up stimulus salience.