Content uploaded by Harold Bekkering
Author content
All content in this area was uploaded by Harold Bekkering
Content may be subject to copyright.
Semantic Activation in Action Planning
Oliver Lindemann
Radboud University Nijmegen and University of Groningen
Prisca Stenneken
Freie Universitaet Berlin
Hein T. van Schie and Harold Bekkering
Radboud University Nijmegen
Four experiments investigated activation of semantic information in action preparation. Participants
either prepared to grasp and use an object (e.g., to drink from a cup) or to lift a finger in association with
the object’s position following a go/no-go lexical-decision task. Word stimuli were consistent to the
action goals of the object use (Experiment 1) or to the finger lifting (Experiment 2). Movement onset
times yielded a double dissociation of consistency effects between action preparation and word process-
ing. This effect was also present for semantic categorizations (Experiment 3), but disappeared when
introducing a letter identification task (Experiment 4). In sum, our findings indicate that action semantics
are activated selectively in accordance with the specific action intention of an actor.
Keywords: semantic action representation, goal-directed action, grasping, tool use
In the area of motor control, many sophisticated models have
been developed during the last couple of decades that specified the
parameters of control for making object-oriented hand movements
(Rosenbaum, 1991). However, a long neglected issue concerns the
role of semantic knowledge in the process of action planning and
control (see Creem & Profitt, 2001). That is, we do not only attune
our motor system to the physical properties of a stimulus, but we
also use our knowledge of what to do with an object and how to
use it.
Recently, several behavioral and neuroimaging studies demon-
strated that the visual perception of graspable objects and prepar-
ing for action are mutually dependent processes. For example, it
has been shown that passive observations of tools evoke neuronal
activation in different cortical motor areas (Martin, Wiggs, Unger-
leider, & Haxby, 1996; Grafton, Fadiga, Arbib, & Rizzolatti, 1997;
Chao & Martin, 2000) and facilitate motor responses that are
consistent with these objects (Tucker & Ellis, 1998; Ellis &
Tucker, 2000). Interestingly, other studies assume effects of a
reversed directionality, that is, they assume effects of action on
perception (e.g., Mu¨sseler & Hommel, 1997; Wohlschla¨ger, 2000;
Mu¨sseler, Steininger, & Wu¨hr, 2001; Creem-Regehr, Gooch,
Sahm, & Thompson, 2004). Several studies, for instance, showed
that the planning or preparation of a motor action is able to
facilitate visual processing, such as the detection of a visual
stimulus that is consistent with the intended action (e.g., Craigh-
ero, Fadiga, Rizzolatti, & Umilta` 1999; Bekkering & Neggers,
2002; Hannus, Cornelissen, Lindemann, & Bekkering, 2005).
Despite the increasing evidence of the direct coupling between
visual perception and action in motor control, the underlying
mechanisms and representations are not well understood (see
Hommel, Mu¨sseler, Aschersleben, & Prinz, 2001 for recent review
and theoretical considerations). In particular, not much is known
about the role of semantic knowledge in action planning. Several
neuropsychological studies have shown that there are patients with
apraxia who have a selective deficit in object use but spared
semantic knowledge about those objects (e.g., Buxbaum,
Schwartz, & Carew, 1997; Rumiati, Zanini, Vorano, & Shallice,
2001). On the other hand, patients have been reported with seman-
tic loss but with the ability to manipulate objects accordingly (e.g.,
Buxbaum et al., 1997; Lauro-Grotto, Piccini, & Shallice, 1997).
This indicates that the two domains, action planning and semantic
knowledge, are at some level independent from each other and that
the accessibility of conceptual knowledge is not necessarily re-
quired for an appropriate object-directed action. Comparable find-
ings led Riddoch, Humphreys, and Price (1989) to conclude that
there is a direct route from vision to action that bypasses seman-
tics. This notion received additional support from experiments
with neuropsychological intact adults (Rumiati & Humphreys,
1998). Interestingly, however, a recent study by Creem and Proffitt
(2001) indicated that action planning and semantic processing
cannot be considered under all circumstances as two independent
processes. They found in a dual-task experiment that normal
subjects often used inappropriate grasping for household tools
when object grasping was paired with a semantic dual task, but less
so when paired with a visuospatial dual task. As the authors
Oliver Lindemann, Nijmegen Institute for Cognition and Information,
Radboud University Nijmegen, the Netherlands, and School of Behavioral
and Cognitive Neuroscience, University of Groningen, the Netherlands;
Prisca Stenneken, General and Experimental Psychology, Freie Universi-
taet Berlin, Germany; Hein T. van Schie and Harold Bekkering, Nijmegen
Institute for Cognition and Information, Radboud University Nijmegen, the
Netherlands.
We acknowledge Karin Roze, Corrine van den Brom, and Boris van
Waterschoot for collecting the data, and we thank Bernhard Hommel for
helpful comments on previous versions of this article.
Correspondence concerning this article should be addressed to Oliver
Lindemann, Nijmegen Institute for Cognition and Information, Radboud
University Nijmegen, P.O. Box 9104, 6500 HE Nijmegen, The Nether-
lands. E-mail: o.lindemann@nici.ru.nl
Journal of Experimental Psychology: Copyright 2006 by the American Psychological Association
Human Perception and Performance
2006, Vol. 32, No. 3, 633– 643
0096-1523/06/$12.00 DOI: 10.1037/0096-1523.32.3.633
633
argued, this finding indicates that semantic processing is involved
when preparing to grasp a meaningful object. The notion of the
important role of functional knowledge in object-directed motor
action is also supported by behavioral and developmental studies
in children and adults indicating that in our everyday life, we build
up strong associations between objects and hand shapes (Klatzky,
Pellegrino, McCloskey, & Doherty, 1989; Klatzky, Pellegrino,
McCloskey, & Lederman, 1993) and the purpose or function for
which objects are typically used (Rosch, Mervis, Gray, Johnson, &
Boyes-Braem, 1976; McGregor, Friedman, Reilly, & Newman,
2002).
The importance of semantics for action is furthermore reflected
by the results of behavioral studies that showed that semantic
properties of distracting words (Gentilucci, Benuzzi, Bertolani,
Daprati, & Gangitano, 2000; Gentilucci & Gangitano, 1998;
Glover & Dixon, 2002; Glover, Rosenbaum, Graham, & Dixon,
2004) or objects (Jervis, Bennett, Thomas, Lim, & Castiello, 1999)
influenced the kinematics of reach-to-grasp movements. For in-
stance, Gentilucci et al. (2000) reported that Italian words denoting
far and near printed on to-be-grasped objects had comparable
effects on movement kinematics as the actual greater or shorter
distances between hand position and object. Glover and Dixon
(2002) reported that maximum grip aperture was enlarged when
subjects grasped an object with the word large printed on top, as
compared to grasping of an object with label small. Another effect
indicating an interaction between semantics and action was re-
ported by Glenberg and Kaschak (2002). They instructed their
participants to judge whether sentences were sensible by making a
motor response that required moving toward or away from their
bodies and found faster response latencies when the sentence
implied an action in same direction (e.g., “Close the drawer,”
which implies an action away from the body) as the direction of
the required motor response (e.g., moving their hand away from
their body to indicate “yes”). According to the authors, this di-
rectly supports the notion that language comprehension is
grounded in bodily actions.
The studies previously mentioned nicely demonstrate the impact
of semantic information on the action system, showing the readi-
ness in which semantic content, for example, from words, may
interfere with and influence ongoing behavioral performance. It is
typically not the case that mere activation of semantic information
will in itself result in the execution of a stereotypical action,
however (with the exception of patients that display utilization
behavior; Archibald, Mateer, & Kerns, 2001). Rather, human
behavior, in unaffected cases, shows the ability to withstand many
of the automatic tendencies or affordances that may be present in
the environment and to control action selection in accordance with
immediate and long-term behavioral goals (Norman & Shallice,
1986; see Humphreys & Riddoch, 2000; Rumiati et al., 2001 for
neuropsychological cases in which there is a deficit in supervisory
attentional control).
Although it is clear that executive processes that regulate the
coherence of goal-directed behavior over time, must at some point,
modulate the influence of action semantics on behavior, the exact
interaction between the two mechanisms remains to be determined.
One possibility is that semantic information on the functional use
of objects is activated automatically upon presentation of those
objects and that the control mechanisms for action subsequently
select the most favorable course of action from the available
alternatives (Buxbaum, 2001). Another possibility is that the acti-
vation of semantic information is selectively modulated in accor-
dance with the behavioral goals of the task that the person is
involved in. In this case, the semantic properties of the object will
not be activated in full, but only those aspects that are relevant for
the ongoing task. This hypothesis would be consistent with a
selection-for-action viewpoint (Allport, 1987) in which informa-
tion, whether perceptual or semantic, is selected in accordance
with the action intention of the person that is about to act. In partial
support for this possibility, electrophysiological studies indicate
that providing subjects with specific task instructions to attend and
respond to certain object properties, determines the type of seman-
tic information that is activated to those objects (Coltheart, Inglis,
Cupples, Michie, Bates, & Budd, 1998).
Whereas interactions between perception and action have been
studied in both directions (effects of perception on action and
influence of action preparation on perception; see information
previously mentioned), there have been hardly any studies that
looked into the influence of action preparation on the level of
semantics. In the present study, we attempt to learn more about the
activation of semantic information in the course of action prepa-
ration and tested the hypothesis that semantic action knowledge is
activated in accordance with the specific action intention of the
actor.
Traditionally, language tasks have been used to investigate
semantics. A typical finding is that the semantic context (e.g.,
provided by a prime word) facilitates the processing of semanti-
cally related words (for review, see Neely, 1991). Priming effects
have often been studied with a lexical-decision task in which
participants have to judge whether a visually presented letter string
is a lexically valid word or not. The semantic priming effect is very
robust and has been supposed to occur automatically (Neely,
1991). It is plausible to assume that semantic preactivation is not
restricted to the linguistic domain (e.g., from prime word to target
word). Semantic effects have been reliably found between linguis-
tic and nonlinguistic stimuli (e.g., Lucas, 2000; Van Schie, Wijers,
Kellenbach, & Stowe, 2003). Additionally, priming studies have
indicated facilitation for a variety of prime-target relations, includ-
ing script relations, functional relations, and perceptual relations
(overview in Lucas, 2000).
To investigate effects of action preparation on semantics, four
experiments were conducted in which the preparation of an action
provided the semantic context for a subsequently presented word.
In all experiments, participants prepared a motor action (e.g., drink
from a cup) and delayed its execution until a word appeared on a
screen. In Experiment 1, participants were required to execute the
action (go) if the word was lexically valid, but withhold from
responding if a pseudoword was presented (no-go). The size of
interference between action preparation and lexical decision was
estimated by comparing the movement onset times in trials with
action-consistent words (e.g., mouth) and action inconsistent
words (e.g., eye). To ensure that the expected action word-
processing interaction depended on the relation between prepared
action and processed words and not on the sequence of the pre-
sented stimuli (i.e., picture-word priming, cf. Vanderwart, 1984;
Bajo & Canas, 1989), a control condition was introduced in which
participants were required to perform simple finger-lifting move-
ments instead of grasping responses. Assuming that the semantic
concepts associated with the goal location of the object use are
634
LINDEMANN, STENNEKEN, VAN SCHIE, AND BEKKERING
only activated with the preparation to grasp the objects, interac-
tions between action planning and word processing were only
expected in the grasping condition. The preparation of finger-
lifting responses, however, should not activate these semantic
concepts.
Experiment 1
The aim of Experiment 1 was to investigate the activation of
action semantics in association with action preparation. We re-
quired our subjects to either grasp and use one of two objects (cup
or magnifying glass) or to lift one of two fingers related to the
object positions. Subsequently presented words in a go/no-go
lexical-decision task (Gordon, 1983) determined whether the pre-
pared motor action should be executed (go) or not (no-go). In line
with the hypothesis that action semantics are activated conform the
action intention of the subjects, we expected faster responses in the
object-grasping condition for trials in which words were consistent
with the goal location of the object use, as compared to trials with
inconsistent words. In contrast, no latency differences between
consistent and inconsistent words were expected for finger-lifting
responses.
Method
Participants. Twenty-four students (18 females and 6 males) from the
University of Nijmegen took part in the experiment. All were right-handed
and Dutch native speakers.
Setup. Figure 1 illustrates the experimental setup. In front of the
participants, we placed a computer display and a touch-sensitive response
box with markers to indicate and control the starting position of the right
hand. Additionally, a cylindrical cup without any handle (diameter 7.5 cm,
height 10.0 cm) and a round magnifying glass (diameter 7.5 cm) with a
handgrip (length 9.0 cm) were situated on the table, both at a reaching
distance of 33 cm. To keep the object positions constant, we used a desk
pad with drawings of the object contours. The object positions (left
side/right side) were counterbalanced between the participants.
Procedure. All participants were randomly assigned to one of two
action conditions (object grasping or finger-lifting). At the beginning of
each trial, a picture of one of the two objects appeared on the screen for 500
ms. In the object-grasping condition, participants were instructed to pre-
pare actions associated with these objects. None of these actions was
described verbally, nor were actions or their endpoints mentioned in the
task instructions. Instead, the experimenter performed the associated ac-
tions in presence of the subject to instruct the required motor responses at
the beginning of the experiment. For example, if a cup was shown, the
required action was to grasp the cup and to bring it to the mouth. The motor
response associated with the magnifying glass was to grasp the object and
move it to the right eye. By contrast, in the finger-lifting condition, the
participants prepared a lifting of either the index or middle finger of the
right hand depending on which side the depicted object was situated on the
table. Importantly, the action in association with the object had to be
delayed until the presentation of a word on the screen. After a variable
delay of 500 ms to 2,000 ms either a valid Dutch word or a pseudoword
was presented for 1,000 ms. We instructed our subjects to initiate the
prepared action as soon as the word was identified as a lexically valid word
(go), and to place back the object after the action was finished. Whenever
a pseudoword was displayed, participants were instructed to withhold from
responding (no-go). In the object-grasping condition, a cross appeared on
the screen 2,500 ms after word offset and extinguished when the subject
returned the hand correctly to the starting position. Because the time
needed to execute a simple finger movement is relatively short, the cross
in the finger-lifting condition was presented 1,000 ms after word offset.
Stimuli and design. The target words used for the go/no-go lexical-
decision task were the Dutch words MOND [mouth] and OOG [eye]
representing the goal locations of the action associated with the cup and the
action associated with the magnifying glass, respectively. In addition, two
unrelated filler words, DEUR [door] and TAS [bag], were selected to match
the target words with respect to word category, word length (three or four
letters, monosyllabic), and word frequency in written Dutch language
(CELEX lexical database, Burnage, 1990). Thus, in go trials, the presented
words were either consistent with respect to the prepared action, inconsis-
tent with the action (e.g., associated with the other object), or unrelated
fillers. Additionally, five legal pseudowords were constructed for the no-go
trials. These were derived from the targets by replacing all letters (vowels
by vowels and consonants by consonants) so that the syllable structure and
the word length were identical to those of targets. All pseudowords obeyed
the Dutch phonotactics.
Thus, there were two action conditions (object grasping and finger
lifting) varied between subjects. Each condition consisted of 96 target trials
(50% action-consistent words, 50% action-inconsistent words), 48 filler
trials, and 30 (17.2%) no-go trials (2 objects ⫻ 5 pseudowords ⫻ 3
repetitions). All trials were presented in a randomized sequence. The
experiment lasted about 30 minutes.
Data acquisition and analysis. To record hand and finger movements,
we used an electromagnetic position tracking system (miniBIRD 800™,
Ascension Technology Corporation). In the object-grasping condition,
three sensors were attached to the participants’ thumb, index finger, and
wrist of their right hand. Only two sensors, one attached to the right index
finger and one to the middle finger, were needed in the finger-lifting
condition. Sensor positions were tracked with a sampling rate of 103.3 Hz.
Movement kinematics were analyzed off-line. We applied a fourth-order
Butterworth low-pass filter with a cutoff frequency of 10 Hz on the raw
data. Two criteria were chosen to detect onsets and offsets of the reach-
to-grasp and finger-lifting movements. An onset was defined to be the first
moment in time when the tangential velocity exceeded the threshold of 10
cm/s and remained above this level for the minimum duration of 400 ms
(object-grasping condition) or 50 ms (finger-lifting condition). For the
offsets, we used the reversed criteria, taking the time of the first sample
where the velocity decreased below the threshold for the predefined time.
The time differences between word onset and hand movement onset
(determined by the wrist sensor) were used to calculate response latencies
in the object-grasping condition. We additionally calculated the following
Figure 1. Illustration of the experimental setup. A cup (Object 1), a
magnifying glass (Object 2), and a computer display were placed on the
table. Participants placed their right hand on the starting position.
635
SEMANTIC ACTIVATION IN ACTION PLANNING
kinematic parameters of the first movement after word presentation: reach
time, peak velocity, and percentage of time to maximum grip (TMG)
aperture with respect to reach time. In the finger-lifting condition, the
analysis was restricted to the response latencies determined by the onset of
the first finger movement after word presentation. All trials with incorrect
responses (i.e., incorrect lexical decisions and wrong actions) or with
response latencies more than 1.5 standard deviations from each partici-
pant’s mean were excluded from the statistical analysis (cf. Ratcliff, 1993).
A Type I error rate of
␣
⫽ .05 was used in all statistical tests reported
in this article. Given
␣
⫽ .05 and n ⫽ 12 participants for each action
condition, contrasts between consistent and inconsistent trials of the size
d
3
⬘⫽.9 (cf. Cohen, 1977) could be detected with a probability of (1-

) ⫽
.81 for object grasping as well as for finger lifting.
1
Results and Discussion
For both action conditions, the percentages of correct lexical
decisions to Dutch words (hits) were greater than 98.4%. No false
alarm responses occurred. Wrong actions (i.e., lifting the wrong
finger or grasping the wrong object) occurred in less than 1% of all
hit trials.
The response latencies to target words (i.e., consistent and
inconsistent words) and filler words did not differ significantly,
neither for object grasping, t(11) ⫽⫺2.00, p ⬎ .05, nor for finger
lifting, |t(11)| ⬍ 1. For further analyses. we focused on the contrast
between consistent and inconsistent trials.
The mean response latencies to target words in the lexical-
decision task are shown in Figure 2. A mixed-model analysis of
variance (ANOVA) with one between-subjects factor (action con-
dition: object grasping or finger lifting) and one within-subject
factor (word consistency) was computed. The data showed an
overall trend for the factor word consistency, F(1, 22) ⫽ 3.91, p ⫽
.06. Most important, the interaction between the two factors was
significant, F(1, 22) ⫽ 4.72, p ⬍ .05. Post hoc two-tailed t tests
showed that for object grasping, the latencies to consistent words
(521 ms) were shorter than to inconsistent words (538 ms), t(11) ⫽
⫺3.35, p ⬍ .01, d
3
⬘⫽1.36 (cf. Cohen, 1977), whereas no
significant difference was observed in the finger-lifting condition
(consistent: 493 ms vs. inconsistent: 492 ms), |t(11)| ⬍ 1.
For the object-grasping condition, we calculated the kinematic
parameters reach time, peak velocity, and the percentage of TMG
aperture. They were entered separately into three 2 (object) ⫻ 2
(word consistency) repeated measures ANOVAs (see Table 1,
Experiment 1). When grasping the cup, peak velocity was slower,
F(1, 11) ⫽ 12.58, p ⬍ .01, and TMG was later, F(1, 11) ⫽ 15.04,
p ⬍ .01, as compared to trials in which the magnifying glass was
grasped. The reach times did not differ significantly. More impor-
tant, no effects of the word consistency were found in any kine-
matic variable, all F ⬍ 1.
In summary, response latencies to words consistent with the
action goal of the movements were faster if object grasping was
prepared. On the contrary, no consistency effects were found when
subjects prepared for finger lifting. This dissociation suggests that
the action word-processing interaction did not merely arise from
presenting the object pictures or from attention being selectively
directed to one of the two objects. Both action conditions were
identical in these aspects. We can also exclude the alternative
explanation that simple picture-word priming (cf. Vanderwart,
1984; Bajo & Canas, 1989) caused the response latency differ-
ences, because pictures and words were identical in both condi-
tions. Rather, the results suggest that action-specific semantic
information was selected in association with the action goal of the
movement that was prepared. Only when subjects had the intention
to grasp and use the objects, a relative advantage was found for
words that specified the goal location of the object use. Further-
more, the absence of priming effects in the finger-lifting condition
argues against the hypothesis that action semantics are activated
automatically and independent from the behavioral goal upon the
presentation of objects.
Nevertheless, there is a possible alternative account for the lack
of response latency differences in the finger-lifting condition. In
Experiment 1, the preparation of simple finger movements appears
much easier than the preparation of reach-to-grasp movements,
which are motorically more complex. As a consequence, partici-
pants may have been more efficient in cognitively separating
action preparation and word recognition tasks, resulting in the
1
All statistical power analyses reported here were performed using the
GPOWER program (Erdfelder, Faul, & Buchner, 1996).
Figure 2. Mean response latencies in the go/no-go lexical-decision task of Experiment 1 (goal locations) and
2 (spatial descriptions) as a function of the factors Action Condition and Word Consistency. Error bars represent
the 95% within-subject confidence intervals (cf. Loftus & Masson, 1994).
636
LINDEMANN, STENNEKEN, VAN SCHIE, AND BEKKERING
absence of an action word-processing interaction for the finger-
lifting condition.
Experiment 2
Experiment 2 was conducted to exclude the possibility that the
differences in the motor complexity in the two action conditions
(object grasping and finger lifting) may have affected interactions
between action preparation and lexical decision. In order to test
that the effects reported in Experiment 1 were independent of
movement complexity and the result of an interaction between the
semantic representations involved in motor preparation and word
processing, we sought to reverse the pattern of effects between the
two conditions. Instead of using words related to the goal locations
of grasping movements, we presented the Dutch words for left and
right, as representatives of action features believed to be important
in the finger-lifting condition. We hypothesized that these spatial
descriptions were much more relevant for the finger-lifting condi-
tion than for the object-grasping condition. In other words, we
predicted an action word-processing interaction primarily for the
finger-lifting condition and much smaller or no effects for object
grasping.
Method
Participants. Again, 24 students (16 females and 8 males) from the
University of Nijmegen were tested. All were right-handed and Dutch
native speakers.
Setup and procedure. The experimental setup and procedure was the
same as compared to Experiment 1.
Stimuli and design. For the go/no-go lexical-decision task, we used as
consistent and inconsistent words (target words) the Dutch words for the
spatial relations, left and right (e.g., LINKS and RECHTS). The unrelated
filler words were BLAUW [blue] and SCHOON [nice, clean], selected to
match target words in word category, word length (i.e., number of letters
and syllables), and word frequency in written Dutch language. For the
no-go trials, five legal pseudowords with the same word lengths and
syllable structure as the target words were constructed from the targets
by replacing all letters. The experimental design was identical to Experi-
ment 1.
Data acquisition and analysis. Data acquisition and analysis was un-
changed. Also, the statistical power to detected consistency effects was
identical to those in Experiment 1.
Results and Discussion
Hit rates for both action conditions were higher than 98.8%. No
false alarm responses occurred. The percentages of wrong actions
were 2.8% for object grasping and 0.8% for finger lifting. In
both action conditions, response latencies to filler words were
statistically not different from the latencies to target words, both
|t(11)| ⬍ 1.
Means response latencies to target words in the lexical-decision
task are shown in Figure 2. The 2 (action condition) ⫻ 2 (word
consistency) mixed-model ANOVA yielded no main effects.
Again, a significant interaction between word consistency and
action condition was found, F(1, 22) ⫽ 10.48, p ⬍ .001. Interest-
ingly, in the finger-lifting condition, response latencies to incon-
sistent words (524 ms) were longer than to consistent words (504
ms), t(11) ⫽ 2.82, p ⬍ .05, d
3
⬘⫽1.15. However, no significant
differences were found in the object grasping condition (consis-
tent: 520 ms vs. inconsistent: 514 ms), t(11) ⫽ 1.58, p ⬎ .05.
The three two-factorial repeated measures ANOVAs (object x
word consistency) of the kinematic parameters revealed that grasp-
ing the cup led to slower peak velocities, F(1, 11) ⫽ 8.43, p ⬍ .05,
and later TMG, F(1, 11) ⫽ 30.02, p ⬍ .001 (see Table 1, Exper-
iment 2). There was no influence of the word consistency on any
kinematic variable, all F(1, 11) ⬍ 1.
In summary, consistent with the results of the previous experi-
ment, Experiment 2 showed an action word-processing interaction
using the spatial descriptions left and right. Again, no effect of the
word consistency was found in the analysis of kinematics. Impor-
tantly, however, in contrast to the first experiment, consistency
effects were now present in the finger-lifting condition, whereas
the response latencies in the object-grasping condition were unaf-
fected by the word meaning. Overall, there is a double dissociation
of effects between Experiment 1 and 2, which indicates that the
action word-processing interaction cannot be explained by the
complexity of the motor response. Rather, the results of Experi-
ments 1 and 2 are consistent with the hypothesis that action
semantics were specifically activated in association with the action
intended by the subject.
Although it is difficult to imagine an explanation for the action
word-processing interaction without including semantics, there is a
possibility that the repeated use of the same words in the experi-
ment may have led participants to perform the lexical-decision task
on the basis of the visual word forms alone, without involving
semantics. To control for this possibility and to provide further
Table 1
Means of Kinematic Parameters Reach Time (RET), Peak
Velocity (PV), and Percentage of Time to Maximum Grip
Aperture With Respect to Reach Time (TMG) in Experiments 1,
2, 3, and 4 as a Function of Word Consistency
Cup Magnifying Glass
CICI
Experiment 1
RET (ms) 592 597 585 582
PV (cm/s) 123.9 123.4 137.1 137.7
TMG (%) 76.5 75.8 51.8 50.6
Experiment 2
RET (ms) 628 635 637 638
PV (cm/s) 128.6 127.6 141.2 143.0
TMG (%) 77.7 78.5 41.0 38.7
Experiment 3
RET (ms) 522 522 517 516
PV (cm/s) 117.6 117.1 125.7 125.1
TMG (%) 76.1 76.5 71.8 71.9
Experiment 4
RET (ms) 508 508 527 523
PV (cm/s) 118.8 118.6 123.5 125.9
TMG (%) 70.1 70.7 59.8 59.6
Note. C: Consistent words; I: Inconsistent words.
637
SEMANTIC ACTIVATION IN ACTION PLANNING
support for the idea that the interaction between action preparation
and word processing critically depends on semantic processing,
two additional experiments were performed, in which we intro-
duced a semantic categorization and a letter identification task.
Experiment 3
In order to better understand the nature of the action word-
processing interaction, we introduced a semantic categorization
task instead of a lexical-decision task for the current experiment.
We cannot exclude the possibility that the participants in the first
two experiments did read the words, but relied only on the visual
word forms and did not process words to a semantic level. Clearly,
a semantic categorization task cannot be performed without deep
semantic processing. Thus, Experiment 3 allows collecting further
evidence for the assumption that the relevant processing level for
the action word-processing interaction is the semantic processing
level.
Method
Participants. Fifteen students (13 females and 2 males) from the
University of Nijmegen participated in Experiment 3 in return for 4€ or
course credits. All were right-handed and Dutch native speakers.
Setup and procedure. Although the experimental setup was the same
as in Experiment 1, instead of using a touch-sensitive response box, we
implemented an online control function in the motiontracking software to
control whether the hand was positioned correctly at the beginning of each
trial. A white and tangible circle (3 cm) on the desk pad served as a marker
for the initial position.
The procedure was basically identical to Experiment 1. That is, each trial
started with a picture of one of the objects, which indicated the action to
prepare. After a variable delay, a word appeared, which triggered the action
initiation. In contrast to the previous experiments, a semantic categoriza-
tion task was used for the go/no-go decisions. Participants were instructed
to decide whether the displayed word represents a human body part or an
animal. In the case of a body part, the prepared action had to be initiated
immediately; in the case of an animal, no response was required. The
semantic categorizations had to be performed as fast and accurate as
possible. Because the aim of this experiment was to replicate the action
word-processing interaction in object grasping, we could refrain from
varying the type of action. Thus, all participants were required to prepare
and execute reach-to-grasp movements to use the objects.
Materials. The 12 Dutch words that were used for the semantic cate-
gorization task are printed in Table 2. As in Experiment 1, the words
MOND [mouth] and OOG [eye] were the target words, which were con-
sistent or inconsistent with respect to the goal location of the prepared
action. Additionally, we used four action unrelated words, all members of
the natural category of the human body as filler words and six members of
the natural category of animals as no-go stimuli. Both categories were part
of the supracategory natural and were chosen so that they were roughly
comparable in category size. All words were selected to match the two
target words with respect to word length (three or four letters and mono-
syllabic), word category, and word frequency in written Dutch language
(cf. CELEX lexical database, Burnage, 1990). Both target words were
repeated 20 times in combination with each of the two object pictures
indicating the respective action. In order to obtain an equal amount of trials
with target words and filler words, the four fillers were presented 10 times
per object. All six no-go stimuli were repeated four times per object.
The experiment consisted of 80 target trials (50% action consistent
words, 50% action inconsistent words), 80 filler trials, and 48 (23.0%)
no-go trials. All trials were presented in a randomized sequence. The
experiment lasted about 40 minutes.
Data acquisition and analysis. Data acquisition and analyses of laten-
cies and kinematics were as described in Experiment 1. Given
␣
⫽ .05 and
n ⫽ 15 participants, consistency effects of size d
3
⬘⫽.8 could be detected
with a probability of (1-

) ⫽ .82.
Results and Discussion
The percentage of correctly categorized body parts (hits) was
98.2%. Incorrect categorizations of animals (false alarms) oc-
curred on average in less than 1% of all trials. Incorrect actions
were observed in less than 1% of the hit trials. Response latencies
to action-unrelated filler words (499 ms) were slower than to target
words (482 ms), t(14) ⫽⫺6.10, p ⬍ .001. This difference is not
surprising, because filler words were presented less frequently than
target stimuli.
Mean response latencies to consistent and inconsistent target
words for the semantic categorization task are shown in Figure 3.
As expected, the responses to action-consistent words (474 ms)
were significantly faster as compared to action-inconsistent words
(490 ms), t(14) ⫽ 2.37, p ⬍ .05, d
3
⬘⫽0.87.
The 2 (object) ⫻ 2 (word consistency) repeated measures
ANOVAs of the reach time, peak velocity, and percentage of TMG
aperture yielded no influence of word meaning on any kinematic
variable, all F(1, 15) ⬍ 1 (see Table 1, Experiment 3). Merely
effects of the objects were present, that is, grasping the cup led to
slower peak velocities, F(1, 15) ⫽ 7.97, p ⬍ .05, and later TMG,
F(1, 15) ⫽ 5.72, p ⬍ .05.
The aim of the present experiment was to provide further
evidence for a semantic nature of the action word-processing
interaction observed in Experiment 1. Unambiguously, the cate-
gorization task in the present experiment required deep semantic
processing, and the low number of errors in this task indicates that
the participants successfully performed semantic processing. Sim-
ilar to the results of Experiment 1, response latencies were faster
for conditions in which the word semantics were consistent with
the prepared action, without any effects in movement kinematics.
The results of Experiment 3 indicate the reliability of the action
word-processing interaction and support our assumption that the
effect reflects an interaction at a semantic level.
Table 2
Word Stimuli Used in Experiment 3 (Semantic Categorization
Task) and Experiment 4 (Final Letter Identification Task)
Stimulus category
Word stimulus
(Dutch)
English
translation
Experiment 3
(SC task)
Experiment 4
(FLI task)
mond Mouth Target Target
oog Eye Target Target
heup Hip Filler No-go
rug Back Filler Filler
nek Nape Filler No-go
buik Stomach Filler No-go
mug Mosquito No-go Filler
mier Ant No-go No-go
eend Duck No-go Filler
kat Cat No-go No-go
vis Fish No-go No-go
hond Dog No-go Filler
638
LINDEMANN, STENNEKEN, VAN SCHIE, AND BEKKERING
Still, one additional test may be applied to strengthen the present
conclusions. If the interaction between action preparation and
word processing critically depends on semantic processing, the
effect should disappear under conditions in which the activation of
word semantics is not required to solve the task.
Experiment 4
Previous studies using a word-to-word priming paradigm have
shown that the semantic priming effect is reduced or eliminated
when participants perform a letter identification task on the prime
word (prime-task effect, e.g., Henik, Friedrich, Tzelgov, &
Tramer, 1994; Stolz & Besner, 1996). Similarly, Stroop interfer-
ence can be reduced or eliminated when only a single letter is
colored instead of the whole word, as in the standard version of the
task (Besner, Stolz, & Boutilier, 1997). In both tasks, allocating
attention to low-level features of the word is assumed to hinder
semantic processing.
In the present experiment, we transferred this logic to the
paradigm involving action preparation and word reading. We used
the same experimental procedure and the identical stimulus set as
in Experiment 3, although, the go/no-go criterion was whether a
given letter was present in the final position of the word form. If
the observed consistency effects require semantic processing, the
response latency differences should disappear or become signifi-
cantly smaller using a low-level letter-identification task.
Method
Participants. Twenty right-handed and Dutch native-speaking students
(14 females and 6 males) from the University of Nijmegen were tested.
Procedure. The experimental setup and procedure was identical to
Experiment 3. The only modification was that the go/no-go decisions were
based on the final letter of the word that was presented. To be precise,
participants were instructed to initiate the prepared action as soon as
possible only if the word ended with either the letter “D” or “G,” and not
to respond if the word ended with any other letter.
Materials. The same 12 Dutch words as in Experiment 3 were used for
the letter identification (see Table 2). Again, the two goal locations of the
actions MOND [mouth] and OOG [eye] served as target words. Four action
unrelated words (filler words) also ended with a “D” or a “G” and served
as go stimuli. The remaining six words, which did not end with “D” or “G,”
served as no-go stimuli. The frequencies of presentation of targets, fillers,
and no-go stimuli were the same as described in Experiment 3. Thus, there
were 80 target trials (50% action-consistent words, 50% action-inconsistent
words), 80 filler trials, and 48 (23.0%) no-go trials. All trials were pre-
sented in a randomized sequence. The experiment lasted about 40 minutes.
Data acquisition and analysis. Data acquisition and analyses of laten-
cies and kinematics were as described in Experiment 1. In order to interpret
a potential nonsignificant result, as hypothesized, an a priori power anal-
ysis was performed. Given
␣
⫽ .05 and (1-

) ⫽ .80, n ⫽ 20 participants
were needed to detect a consistency effect with somewhat smaller size (d
3
⬘
⫽ .7) than the effects in the first three experiments.
Results and Discussion
The percentage of correct identifications of the final letter “D”
and “G” (hits) was greater than 98%. False alarms occurred in less
than 1% of all trials, and wrong action occurred in less than 1% of
the hit trials. Response latencies to filler words (475 ms) were
longer than to targets words (459 ms), t(19) ⫽ 5.33, p ⬍ .001,
which reflects the fact that filler words were presented less often
than target words.
Mean response latencies to target words for the final letter
identification task are shown in Figure 3. Importantly, there was no
statistical difference between the response latencies to consistent
words (457 ms) as compared to inconsistent words (462 ms),
t(19) ⫽ 1.29.
Again, the three kinematic parameters were analyzed with sep-
arate 2 (object) ⫻ 2 (word consistency) ANOVAs (see Table 1,
Experiment 4). Grasping of the cup led to later TMG, F(1, 19) ⫽
6.80, p ⬍ .05. No effects were observed in peak velocities and
reach times. As in the three experiments before, we did not find
effects of the word meaning on any kinematic variable, all F(1, 19)
⬍ 1.4.
In summary, no reliable action word-processing interaction was
observed with the letter identification task. There were no signif-
icant differences in response latencies for action-consistent words
as compared to inconsistent words. Because the statistical power
was satisfactory, we can exclude the presence of an interaction
between action intention and word semantics in the present exper-
iment. These results are in line with our assumption that when
semantic processing is not required for the go/no-go task, no action
word-processing interaction can be observed. Although some de-
gree of automatic semantic processing of word forms cannot be
excluded in the letter identification task, when attention was di-
rected to low-level features of the target words, the interaction
effects disappeared. Accordingly, the activation of semantic rep-
resentations from the visual word form seems to be a prerequisite
for the observed interaction, which emphasizes the semantic nature
of the action word-processing interaction effect.
General Discussion
The results of the present study demonstrate an interaction effect
between processes involved in action preparation and processes
involved in word reading. Response latencies were sped up if
words presented in a go/no-go decision task were consistent with
Figure 3. Mean response latencies in Experiment 3 (semantic categori-
zation task) and in Experiment 4 (final letter identification task) as a
function of the factors Action Condition and Word Consistency. Error bars
represent the 95% within-subject confidence intervals (cf. Loftus & Mas-
son, 1994).
639
SEMANTIC ACTIVATION IN ACTION PLANNING
the features of a concurrently prepared action. In Experiment 1,
when subjects prepared to grasp the objects, reaction times were
faster when words consistently described the goal location (i.e.,
mouth or cup) of the prepared action. When subjects prepared
finger-lifting movements on the basis of the object positions in-
stead of performing object-directed actions, however, reaction
time effects were found to disappear. These results suggest that
functional semantic information regarding the purpose or action
goal for which an object is typically used does not become acti-
vated automatically upon presentation of the object, but only when
subjects intend to use the object with that specific purpose.
Experiment 2 further supported the hypothesis that semantics
are activated in association with the action intention of the actor
and ruled out possible alternative explanations for the difference
between the two action conditions in Experiment 1. Changing the
words to describe relevant action features for the finger-lifting
condition (left and right) resulted in an action word-processing
interaction for this condition, whereas no effect was found in the
condition in which subjects grasped and used objects. Both the
dissociations between conditions within each experiment and the
reversal of effects between experiments are consistent with our
hypothesis that action semantics about objects are selectively
activated and depend on the actor’s intention.
Experiments 3 and 4 further supported the suggestion that the
action word-processing interaction critically depends on the depth
of semantic processing required by the go/no-go task. In Experi-
ment 3, in which subjects made semantic decisions about word
category instead of the lexical decisions, the action word-
processing interaction between the two tasks was found un-
changed. As Experiment 4 demonstrates, however, the mere pre-
sentation of a visual word form is obviously not sufficient to cause
this effect, which indicates the semantic nature of this effect.
Therefore, we consider the use of a secondary language task as a
successful approach to investigate semantic action representations
and the use of functional object knowledge in the context of action
preparation.
In line with the results of Experiments 3 and 4, which show that
the activation of semantic concepts was critically involved in
establishing the interaction between the two tasks, contemporary
models of motor control (Rosenbaum, Meulenbroek, Vaughan, &
Jansen, 2001; Glover, 2004) suggest that conceptual knowledge is
involved in the selection of appropriate action plans. Consistent
with this notion, recent experiments in the field of motor control
demonstrated that presentation of irrelevant semantic information
(via words) has a direct impact on movement kinematics of reach-
to-grasp actions (Gentilucci & Gangitano, 1998; Gentilucci et al.,
2000; Glover & Dixon, 2002; Glover, 2004). For example, Glover
and Dixon (2002) reported that maximum grip aperture was en-
larged while grasping an object with the word label large as
compared to grasping an object with the word label small. Inter-
estingly, Glover et al., (2004) reported an automatic effect of word
reading on grasp aperture using words for objects that either afford
a large grip (e.g., apple) or a small grip (e.g., grape). In both
studies, Glover and colleagues performed a detailed analysis of the
movement kinematics, which showed that word semantics affected
the motor action only very early in the movement. As the hand
approached the target, the impact of word semantics was found to
decline continuously. In line with these results, the authors con-
cluded that semantic information interferes only with motor plan-
ning but not with motor control once the action is initiated. This
view is consistent with the present findings in which word mean-
ing only affected reaction times but not the online control of
movement execution. Concerning the absence of kinematic effects,
our study notably differs from earlier studies on semantics in
motor action. To be precise, in our paradigm, it was required to
prepare the action before word onset and to execute it after word
processing. In other words, motor action and word reading did not
take place at the same time. This may explain the absence of
kinematics effects in the present study and suggests that the
reaction time differences reflect effects in the word-processing
performances caused by action preparation.
Although the absence of kinematic effects is consistent with the
assumption that actions were prepared effectively, an alternative
account that may partly explain our results is that, instead of
preparing a motor response, participants represented the upcoming
motor task verbally in short-term memory and recalled these
verbal descriptions for the grasping or finger-lifting actions after
picture onset. It must be emphasized, however, that in the present
study, subjects were never instructed in words such as “grasp the
left object and bring it to the mouth.” Instead, subjects just saw the
relevant actions once before the experiment started. Furthermore,
the short reaction times, minimal error rates, and absence of
kinematic effects in all experiments indicate that subjects prepared
the upcoming actions well before word onset and do not suggest
that participants memorized the motor task verbally and prepared
the action only after word onset. This assumption is also supported
by several studies on motor control (e.g., Rosenbaum, 1983; Leu-
thold, Sommer, & Ulrich, 2004), which indicate that, in delayed
response conditions, subjects tend to prepare the motor response as
far as possible in advance instead of maintaining cueing informa-
tion in memory or recalling the task instruction. In light of these
arguments, we consider the verbal working memory explanation to
be unlikely, and we favor the interpretation that effects reflect
semantic overlap between action preparation and lexical
semantics.
Hommel and Mu¨sseler (in press) recently investigated the ef-
fects of action preparation on the perception of directional words
(i.e., left and right). They required their participants to prepare
either a manual left-right keypress response or to say “left” or
“right.” Later, they briefly presented a directional word. In contrast
to the present paradigm, the words had to be recalled after exe-
cuting the prepared response. Under these conditions, planning
vocal actions impaired the perception of directional words, but,
interestingly, the preparation of manual responses did not affect
word processing. Although it might be difficult to compare accu-
racy effects in an unspeeded identification task with reaction time
effects in a lexical-decision task, these results seem to be in
contradiction with the results of the present study. One possible
explanation for this discrepancy is that in the study of Hommel and
Mu¨sseler (in press), participants were required to maintain a short-
term memory representation of the presented words, which was not
the case in the present study. According to the feature-integration
approach (Stoet & Hommel, 1999; Hommel, 2004), attending a
perceptual object as well as planning an action implies an integra-
tion of several activated feature codes into one coherent object
representation or action plan. The mere activation of feature codes
should facilitate processing of all events sharing these features.
Once a feature code is integrated into an action plan or object
640
LINDEMANN, STENNEKEN, VAN SCHIE, AND BEKKERING
representation, however, it is no longer available for another
integration if needed for other cognitive processes. As a result, this
process is assumed to be impaired. It has been suggested that the
likelihood that a feature code becomes integrated depends on the
relevance of the respective feature for the task. That is, unattended
or task-irrelevant features may become activated but not integrated
into one or more bindings (Hommel, 2004). In line of this reason-
ing, in the study of Hommel and Mu¨sseler (Experiment 3A, in
press), feature integration of the word semantics was required to
maintain a short-term memory representation. In the present study,
however, semantic features were activated but were not integrated
into a short-term memory representation, which had to be main-
tained while acting. Consequently, we found that the semantic
congruency between the two tasks did not result in an inhibitory
effect, but in a facilitation of word processing.
Several studies have shown that action preparation as well as
action execution can influence visual perception (Craighero et al.,
1999; Mu¨sseler & Hommel, 1997; Mu¨sseler et al., 2001;
Wohlschla¨ger, 2000). For example, Craighero et al. (1999) re-
quired their participants to prepare an action (i.e., to grasp a bar
with a specific orientation) but to delay action execution until the
appearance of a visual stimulus. Interestingly, they found that
movements were initiated faster when the orientation of the go
stimulus was consistent with the orientation of the to-be-grasped
object. In the same study, consistency effects were found when
after preparation of a hand movement, participants were instructed
to inhibit the prepared grasping movement and to respond with a
different motor effector. Craighero et al. (1999) concluded that the
mere preparation of an action is capable to facilitate the processing
of visual stimuli if it contains features that are consistent with the
preactivated action plan.
In addition to and consistent with the reported perceptual effects
of action preparation, the current study suggests that the principles
of selection for action are also operational at the level of semantics.
Our data suggest that functional semantic information about ob-
jects is activated in association with the action intention of the
subject. For example, although cups are typically brought to the
mouth for drinking, we assume that the concept of mouth is
activated stronger when there is the intention to bring the object to
the mouth as when some other response is required. In fact, our
results suggest that the goal location of the object use is not
activated when there is no specific intention to interact with the
object. That is, when subjects received the instruction to perform
finger-lifting movements, the facilitation of words, which are
consistent with the goal locations of the object use, was absent.
These results suggest that functional semantic information about
how to respond to objects requires motor processing in order to
become activated. A theoretically comparable idea in the field of
perception and action is expressed by the premotor theory of visual
attention (Rizzolatti, Riggio, & Sheliga, 1994), which suggests that
allocating visual attention in space involves covert activation of
the eye movement system. In this perspective, action preparation
no longer just facilitates the selection of perceptual information,
but perceptual processes themselves require motor support. A
similar mechanism might, in theory, be applicable also to the
activation of semantic representations. That is, functional semantic
information about the use of objects may require activation of
motor representations to enable the semantic information to be
addressed. Similar proposals have been made with respect to visual
semantic knowledge about the appearance of objects (Martin,
Haxby, Lalonde, Wiggs, & Ungerleider, 1995; van Schie, Wijers,
Mars, Benjamins, & Stowe, 2005). Lesions in visual areas of the
brain (occipitotemporal area) sometimes not only result in percep-
tual deficits, but may also impair the activation of knowledge
about the visual properties of objects (review in Humphreys &
Forde, 2001). Comparable ideas have been expressed with respect
to functional semantic knowledge in which lesions in motor areas
of the brain are held responsible for subjects’ impairments to
represent the functional properties of objects (review in Saffran &
Schwartz, 1994). A growing number of neuroimaging studies
furthermore affirm the idea that visual and motor representations
support semantic knowledge about the use and appearance of
objects (e.g., Pulvermu¨ller, 1999; Tranel, Kemmerer, Adolphs,
Damasio, & Damasio, 2003). One advance of the present study is
that it provides behavioral support for existing neuropsychological
and neuroimaging results, which suggest that accessing functional
semantic information about objects involves an activation of spe-
cific motor representations.
Whereas an increasing amount of research is directed at the use
of motor representations for semantics, the reverse relationship
concerning the use of semantic information for actions has re-
ceived much less attention. As a consequence, the contribution of
semantics for use of objects is still not fully understood (see
Rumiati & Humphreys, 1998; Creem & Proffitt, 2001; Buxbaum et
al., 1997). For example, neuropsychological studies (e.g., Bux-
baum et al., 1997; Riddoch et al., 1989) and behavioral studies
with time-limited conditions (Rumiati & Humphreys, 1998) sug-
gest that semantic knowledge can be bypassed when selecting an
action as response to objects. This shows that an involvement of
semantics in action planning is not obligatory and has led Riddoch
et al. (1989) to conclude that a direct route from vision to action
exists in addition to a semantic route. Nevertheless, some experi-
mental findings support information-processing models for action,
which propose that access to stored semantic knowledge about an
object is utilized to generate correct object-directed actions
(MacKay, 1985). For example, a recent study of Creem and
Proffitt (2001) shows that semantic processing is required when
grasping ordinary household tools appropriately for their use. They
observed that subjects’ grasping was frequently inappropriate
when the motor task was paired with an additional semantic task
but not when paired with a visuospatial task. In congruence with
these findings, the present study demonstrates the important role
for functional semantic knowledge in action preparation and pro-
vides evidence for the notion that action semantics are routinely
activated with the preparation and execution of goal-directed
actions.
In addition to the rather general conclusion that semantics are
involved in object use, the results of the present study clearly
indicate that action semantics are activated selectively in accor-
dance with the actor’s specific intention. In other words, depend-
ing on the person’s current behavioral goal, different functional
object properties become relevant and activate as a result of
different aspects of semantic action knowledge. This conclusion
does not contradict the results of neuroimaging experiments that
find motor areas activated to the presentation of manipulable
objects (Chao & Martin, 2000; Creem-Regehr & Lee, 2005) or the
findings of behavioral studies suggesting that the mere perception
of tools automatically potentiates components of the actions they
641
SEMANTIC ACTIVATION IN ACTION PLANNING
afford (Tucker & Ellis, 1998; Ellis & Tucker, 2000). In contrast to
these studies, however, the present study points out the modulating
role of action intentions on the activation of action knowledge
related to an object in a functional-relevant way.
In conclusion, the advance that is made with the present para-
digm is that we were able to establish a measure of semantic action
knowledge as it is activated in the process of action preparation.
Our finding of an action word-processing interaction suggests that
the selection-for-action hypothesis is not just restricted to the
domain of perception and action, but it is to be extended to the
field of semantics. This insight certainly calls for further investi-
gation. More scientific interest into the area of semantic action
knowledge is expected to increase our understanding of the cog-
nitive mechanisms that underlie the planning and control of motor
actions.
References
Allport, D. A. (1987). Selection for action: Some behavioral and neuro-
physiological considerations of attention and action. In H. Heuer & A. F.
Sanders (Eds.). Perspectives on perception and action (pp. 395– 419).
Hillsdale, NJ: Erlbaum.
Archibald, S. J., Mateer, C. A., & Kerns, K. A. (2001). Utilization behav-
ior: Clinical manifestations and neurological mechanisms. Neuropsy-
chology Review, 11(3), 117–130.
Bajo, M.-T., & Canas, J.-J. (1989). Phonetic and semantic activation during
picture and word naming. Acta Psychologica, 72(2), 105–115.
Bekkering, H., & Neggers, S. F. W. (2002). Visual search is modulated by
action intentions. Psychological Science, 13(4), 370 –374.
Besner, D., Stolz, J.-A., & Boutilier, C. (1997). The Stroop effect and the
myth of automaticity. Psychonomic Bulletin and Review, 4(2), 221–225.
Burnage, G. (1990). CELEX: A guide for users. Center for Lexical Infor-
mation, University of Nijmegen, The Netherlands.
Buxbaum, L. J. (2001). Ideomotor apraxia: A call to action. Neurocase, 7,
445– 458.
Buxbaum, L. J., Schwartz, M. F., & Carew, T. G. (1997). The role of
semantic memory in object use. Cognitive Neuropsychology, 14(2),
219 –254.
Chao, L. L., & Martin, A. (2000). Representation of manipulable man-
made objects in the dorsal stream. NeuroImage, 12(4), 478– 484.
Cohen, J. (1977). Statistical power analysis for the behavioral sciences
(Rev. ed.). Hillsdale, NJ, England: Erlbaum.
Coltheart, M., Inglis, L., Cupples, L., Michie, P., Bates, A., & Budd, B.
(1998). A semantic system specific to the storage of information about
the visual attributes of animate and nanimate objects. Neurocase, 4,
353–370.
Craighero, L., Fadiga, L., Rizzolatti, G., & Umilta`, C. (1999). Action for
perception: A motor-visual attentional effect. Journal of Experimental
Psychology: Human Perception and Performance, 25(6), 1673–1692.
Creem, S. H., & Proffitt, D. R. (2001). Grasping objects by their handles:
A necessary interaction between cognition and action. Journal of Ex-
perimental Psychology: Human Perception and Performance, 27(1),
218 –228.
Creem-Regehr, S. H., Gooch, A. A., Sahm, C. S., & Thompson, W. B.
(2004). Perceiving virtual geographical slant: Action influences percep-
tion. Journal of Experimental Psychology: Human Perception and Per-
formance, 30(5), 811– 821.
Creem-Regehr, S. H., & Lee, J. N. (2005). Neural representations of
graspable objects: Are tools special? Cognitive Brain Research, 22(3),
457– 469.
Ellis, R., & Tucker, M. (2000). Micro-affordance: The potentiation of
components of action by seen objects. British Journal of Psychology,
91(4), 451– 471.
Erdfelder, E., Faul, F., & Buchner, A. (1996). GPOWER: A general power
analysis program. Behavior Research Methods, Instruments and Com-
puters, 28(1), 1–11.
Gentilucci, M., Benuzzi, F., Bertolani, L., Daprati, E., & Gangitano, M.
(2000). Language and motor control. Experimental Brain Research,
133(4), 468 –490.
Gentilucci, M., & Gangitano, M. (1998). Influence of automatic word
reading on motor control. European Journal of Neuroscience, 10(2),
752–756.
Glenberg, A.-M., & Kaschak, M.-P. (2002). Grounding language in action.
Psychonomic Bulletin and Review, 9(3), 558 –565.
Glover, S. (2004). Separate visual representations in the planning and
control of action. Behavioral and Brain Sciences, 27(1), 3–24.
Glover, S., & Dixon, P. (2002). Semantics affect the planning but not
control of grasping. Experimental Brain Research, 146(3), 383–387.
Glover, S., Rosenbaum, D. A., Graham, J., & Dixon, P. (2004). Grasping
the meaning of words. Experimental Brain Research, 154(1), 103–108.
Gordon, B. (1983). Lexical access and lexical decision: Mechanisms of
frequency sensitivity. Journal of Verbal Learning and Verbal Behavior,
22(1), 24 –44.
Grafton, S. T., Fadiga, L., Arbib, M. A., & Rizzolatti, G. (1997). Premotor
cortex activation during observation and naming of familiar tools. Neu-
roImage, 6(4), 231–236.
Hannus, A., Cornelissen, F. W., Lindemann, O., & Bekkering, H. (2005).
Selection-for-action in visual search. Acta Psychologica, 118, 171–191.
Henik, A., Friedrich, F.-J., Tzelgov, J., & Tramer, S. (1994). Capacity
demands of automatic processes in semantic priming. Memory & Cog-
nition, 22(2), 157–168.
Hommel, B. (2004). Event files: Feature binding in and across perception
and action. Trends in Cognitive Sciences, 8(11), 494 –500.
Hommel, B., & Mu¨sseler, J. (in press). Action-feature integration blinds to
feature-overlapping perceptual events: Evidence from manual and vocal
actions. Quarterly Journal of Experimental Psychology (A).
Hommel, B., Mu¨sseler, J., Aschersleben, G., & Prinz, W. (2001). The
theory of event coding (TEC): A framework for perception and action
planning. Behavioral and Brain Sciences, 24(5), 849–937.
Humphreys, G. W., & Forde, E. M. E. (2001). Hierarchies, similarity, and
interactivity in object recognition: “Category-specific” neuropsycholog-
ical deficits. Behavioral & Brain Sciences, 24, 453–509.
Humphreys, G. W., & Riddoch, J. M. (2000). One more cup of coffee for
the road: Object-action assemblies, response blocking and response
capture after frontal lobe damage. Experimental Brain Research, 133,
81–93.
Jervis, C., Bennett, K., Thomas, J., Lim, S., & Castiello, U. (1999).
Semantic category interference effects upon the reach-to-grasp move-
ment. Neuropsychologia, 37(7), 857– 868.
Klatzky, R. L., Pellegrino, J., McCloskey, B. P., & Lederman, S. J. (1993).
Cognitive representations of functional interactions with objects. Mem-
ory & Cognition, 21(3), 294–303.
Klatzky, R. L., Pellegrino, J. W., McCloskey, B. P., & Doherty, S. (1989).
Can you squeeze a tomato? The role of motor representations in seman-
tic sensibility judgments. Journal of Memory and Language, 28(1),
56 –77.
Lauro-Grotto, R., Piccini, C., & Shallice, T. (1997). Modality-specific
operations in semantic dementia. Cortex, 33(4), 593–622.
Leuthold, H., Sommer, W., & Ulrich, R. (2004). Preparing for action:
Inferences from CNV and LRP. Journal of Psychophysiology, 18(2–3),
77– 88.
Loftus, G. R., & Masson, M. E. J. (1994). Using confidence intervals in
within-subject designs. Psychonomic Bulletin and Review, 1(4), 476 –
490.
Lucas, M. (2000). Semantic priming without association: A meta-analytic
review. Psychonomic Bulletin and Review, 7(4), 618– 630.
MacKay, D. G. (1985). A theory of representation, organization and timing
642
LINDEMANN, STENNEKEN, VAN SCHIE, AND BEKKERING
of actions with implications for sequencing disorders. In E. A. Roy (Ed.),
Neuropsychological studies of apraxia and related disorders (pp. 267–
308). Amsterdam: North-Holland.
Martin, A., Haxby, J. V., Lalonde, F. M., Wiggs, C. L., & Ungerleider,
L. G. (1995). Discrete cortical regions associated with knowledge of
color and knowledge of action. Science, 270, 102–105.
Martin, A., Wiggs, C. L., Ungerleider, L. G., & Haxby, J. V. (1996). Neural
correlates of category-specific knowledge. Nature, 379, 649– 652.
McGregor, K. K., Friedman, R. M., Reilly, R. M., & Newman, R. M.
(2002). Semantic representation and naming in young children. Journal
of Speech, Language, and Hearing Research, 45, 332–346.
Mu¨sseler, J., & Hommel, B. (1997). Blindness to response-compatible
stimuli. Journal of Experimental Psychology: Human Perception and
Performance, 23(3), 861– 872.
Mu¨sseler, J., Steininger, S., & Wu¨hr, P. (2001). Can actions affect percep-
tual processing? Quarterly Journal of Experimental Psychology: Human
Experimental Psychology, 54A(1), 137–154.
Neely, J. H. (1991). Semantic priming effects in visual word recognition:
A selective review of currents findings and theories. In D. Besner & G.
Humphreys (Eds.), Basic processes in reading: Visual word recognition
(pp. 264 –336). Hillsdale, Hove, London: Erlbaum.
Norman, D. A., & Shallice, T. (1986). Attention to action: Willed and
automatic control of behavior. In R. Davidson, G. Schwartz & D.
Shapiro (Eds.), Consciousness and self regulation: Advances in research
and theory (Vol. 4, pp. 1–18). New York: Plenum Press.
Pulvermu¨ller, F. (1999). Words in the brain’s language. Cognitive Neuro-
science, 22(2), 253–336.
Ratcliff, R. (1993). Methods for dealing with reaction time outliers. Psy-
chological Bulletin, 114(3), 510 –532.
Riddoch, M. J., Humphreys, G. W., & Price, C. J. (1989). Routes to action:
Evidence from apraxia. Cognitive Neuropsychology, 6(5), 437–454.
Rizzolatti, G., Riggio, L., & Sheliga, B. (1994). Space and selective
attention. In C. Umilta` & M. Moscovitch (Eds.), Attention and perfor-
mance (Vol. XV, pp. 231–265). Cambridge, MA: MIT Press.
Rosch, E., Mervis, C. B., Gray, W. D., Johnson, D. M., & Boyes-Braem,
P. (1976). Basic objects in natural categories. Cognitive Psychology, 8,
382– 439.
Rosenbaum, D. A. (1983). The movement precuing technique: Assump-
tions, applications and extensions. In R. A. Magill (Ed.), Memory and
control of movement (pp. 231–272). Amsterdam: North-Holland.
Rosenbaum, D. A. (1991). Human motor control. San Diego: Academic
Press.
Rosenbaum, D. A., Meulenbroek, R. J., Vaughan, J., & Jansen, C. (2001).
Posture-based motion planning: Applications to grasping. Psychological
Review, 108(4), 709 –734.
Rumiati, R. I., & Humphreys, G. W. (1998). Recognition by action:
Dissociating visual and semantic routes to action in normal observers.
Journal of Experimental Psychology: Human Perception and Perfor-
mance, 24(2), 631– 647.
Rumiati, R. I., Zanini, S., Vorano, L., & Shallice, T. (2001). A form of
ideational apraxia as a delective deficit of contention scheduling. Cog-
nitive Neuropsychology, 18, 617– 642.
Saffran, E. M., & Schwartz, M. F. (1994). Of cabbages and things:
Semantic memory from a neuropsychological perspective—A tutorial
Review. In C. Umilta` & M. Moscovitch (Eds.). Attention and perfor-
mance (Vol. XV, pp. 507–536). Cambridge, MA: MIT Press.
Stoet, G., & Hommel, B. (1999). Action planning and the temporal binding
of response codes. Journal of Experimental Psychology: Human Per-
ception and Performance, 25, 1625–1640.
Stolz, J.-A., & Besner, D. (1996). Role of set in visual word recognition:
Activation and activation blocking as nonautomatic processes. Journal
of Experimental Psychology: Human Perception and Performance,
22(5), 1166 –1177.
Tranel, D., Kemmerer, D., Adolphs, R., Damasio, H., & Damasio, A. R.
(2003). Neural correlates of conceptual knowledge for actions. Cognitive
Neuropsychology, 20(3), 409 –432.
Tucker, M., & Ellis, R. (1998). On the relations between seen objects and
components of potential actions. Journal of Experimental Psychology:
Human Perception and Performance, 24(3), 830 – 846.
Vanderwart, M. (1984). Priming by pictures in lexical decision. Journal of
Verbal Learning and Verbal Behavior, 23(1), 67–83.
van Schie, H. T., Wijers, A. A., Kellenbach, M. L., & Stowe, L. A. (2003).
An event-related potential investigation of the relationship between
semantic and perceptual levels of representation. Brain and Language,
86, 300 –325.
van Schie, H. T., Wijers, A. A., Mars, R. B., Benjamins, J. S., & Stowe,
L. A. (2005). Processing of visual semantic information to concrete
words: Temporal dynamics and neural mechanisms indicated by event-
related brain potentials. Cognitive Neuropsychology, 22, 364–386.
Wohlschla¨ger, A. (2000). Visual motion priming by invisible actions.
Vision Research, 40, 925–930.
Received January 13, 2005
Revision received November 9, 2005
Accepted November 25, 2005 䡲
643
SEMANTIC ACTIVATION IN ACTION PLANNING
A preview of this full-text is provided by American Psychological Association.
Content available from Journal of Experimental Psychology Human Perception & Performance
This content is subject to copyright. Terms and conditions apply.