Poster presentation at CogSci 2005.

1. Are Abstract Concepts Structured via More Concrete Concepts?
. John L. Dennis (johnlmdennis@mail.utexas.edu) & Arthur B. Markman (markman@psy.utexas.edu)
Psychology Department University of Texas at Austin
Temporally Ambiguous Question
Wednesday Wednesday
Monday Friday Monday Friday
Metaphoric Structuring View
Wednesday
Monday Friday
Wednesday
Monday Friday
Implied Representation of Agency
EMM TMM
past future past future
Relatively Active Relatively Passive
Spatial Analog of EMM Spatial Analog of TMM
Relatively Active Relatively Passive
Experiment 1
Materials
Fillmore (1968) observed that interpersonal action verbs
(IAVs), like ‘kissed’, have an implicit causality that helps us
formulate inferences about causes of action.
Example:
Agency vs. Passivity
Mary kissed Tom.
Mary is the agent and is typically interpreted as the cause of
the action, while Tom is the patient and is typically
interpreted as the recipient.
In order to prime subjects we asked them to unscramble
sentences, where the sentences were constructed with either
the first person subject pronoun, "I" or the first person object
pronoun, "me".
Examples:
Agency
Mary I bridge under kissed the
Passivity
Mary me kissed the bridge under
We hypothesized that the grammatical subject/object of the
SvO sentence construction and the IAVs would prime
representations of agency for those subjects who
unscrambled "I" sentences and passivity for those subjects
who unscrambled "me" sentences.
Introduction
Cognitive Scientists have long debated the connection
between abstract concepts like happiness integrity, and time,
and experiential concepts, like moving, eating, and physical
space (Barsalou, 1999; Clark, 2001; Prinz, 2002).
In this poster we focus on space/time metaphors, which
have been the subject of much previous research on the role
of metaphor in conceptual understanding (Boroditsky, 2000;
McGlone & Harding, 1998) In particular, there are two
distinct metaphors in English for describing the passage of
time: the ego-moving and time-moving metaphors. In this
poster, we point out that these space/time metaphors have an
implied agency within the grammar of the metaphors. We
present 2 studies, one that demonstrates the degree to which
the concept of agency or passivity is active influences the
interpretation of temporal metaphors, and a second that
demonstrates the degree to which abstract representations of
agency influence our thinking about the concrete domain of
space.
Motivation & Background
Linguistic and Psychological Evidence
Metaphoric Representation View (Lakoff & Johnson, 1980)
Abstract concepts are represented via a metaphor from more
concrete concepts.
Perceptual Symbol Systems (Barsalou, 1997)
Actions like predicates and recursion are done by perception.
Embodied Cognition (Clark, 2001; Prinz & Barsalou, 2000)
Representations as context sensitive, embodied, and
temporally based.
Metaphoric Structuring View (Boroditsky, 2000)
Metaphors provide relational structure for abstract concepts,
e.g., the ego-moving and time-moving metaphors.
Ego & Time-Moving Metaphor
Ego-moving metaphor (EMM)
We passed the deadline two days ago.
past deadline future
Time-moving metaphor (TMM)
The deadline passed [us] two days ago.
past deadline future
Metaphoric Structuring View
Boroditsky (2000) argues that the abstract domain of time is
structured via a metaphor to the concrete domain of space.
Spatial Analog of EMM Spatial Analog of TMM
Boroditsky, 2000 Figure 3
Spatially Ambiguous Target Question
Results
As predicted, 24/30 “I” primed subjects selected the
“distant” widget, while 16/30 of the “me” primed subjects
selected the “closer widget.
A Pearson Chi-Square confirmed that this difference was
significant (1, N=60) = 7.18, P < .007. 16/30 control
participants who had not participated in the sentence
unscrambling primes selected the “distant” widget.
Conclusions
One’s implicit representation of agency alters the way that
one thinks about time and space. We demonstrated that the
implicit representation of agency is fundamental to the
space/time metaphor. Finally, we established that one’s
representation of relative agency significantly alters the
perception of a spatially and temporally ambiguous
scenarios.
Discussion
Previous research on the space/time metaphor has focused on
the explicit connection between space and time by asking
one of two related questions: Are there distinct schemas for
the space/metaphors? Is the abstract domain of time
structured in terms of the more concrete and experiential
domain of space? Previous researchers have answered these
questions in the affirmative. We, on the contrary, argue that
the results from this research is a derivative of one’s
unconscious representations of agency/passivity. Since it
seems plausible that agency is incorporated in our
representations of goals, motivations, affect, and self-
regulation, we believe that these factors could also be
important as an alternative explanation of the findings on the
space/time metaphor. Our current research is untangling
these issues.
Design
The experiment design was between subjects, with subjects
either running in either the "I" (“agency”) or the "me”
(“passivity”) priming condition. Immediately following the
primes subjects answered the temporally ambiguous target
question: “Next Wednesday's meeting has been moved
forward two days. What day is the meeting now that it has
been rescheduled?”
Procedure
Subjects were run solely on Wednesdays, in order to
maintain the same reference point for answering the
temporally ambiguous question. Subjects completed three
pages of scrambled sentences in about 20 minutes. Only the
target question was presented for those in the control group.
Predictions
We predicted that one’s implicit representation of agency
alters how one thinks about time such that the ‘agency’ and
‘passivity’ primes will induce subjects to answer the
temporally ambiguous ‘next Wednesday’s meeting’
question consistent with the ego-moving metaphor and the
time-moving metaphor respectively.
Results
As predicted, for the “I” prime 30/42 selected Friday, the
ego-moving frame of reference, while for the “me” prime
25/42 selected Monday, the time-moving frame of
reference.
A Pearson Chi-Square confirmed that this difference was
significant, (1, N=84) = 8.16, P < .004. Control participants
who had not participated in the sentence unscrambling
primes were about evenly split between Monday (19/42)
and Friday.
Experiment 2
Materials
Experiment 2 was designed to test whether one’s abstract
representation of agency can alter one’s thinking of a more
concrete or experiential representation - i.e., space.
Design
The design of this experiment was exactly the same as
experiment 1 except immediately following the sentence
unscrambling subjects were asked to answer the spatially
ambiguous target question as used by Boroditsky (2000).
Predictions
We predicted that one’s implicit
representation of agency will alter how
one thinks about space thus allowing
subjects to disambiguate the scenario.
such that the ‘agency’ primed subjects
will select the ‘distant' widget while the
‘passivity’ primes will select the ‘closer’
widget.
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0.90
1.00
Control (No Prime) Agency ("I" Grammatical
Subject)
Passivity ("ME"
Grammatical Object)
Condition
Proportion Selecting Friday
Ego-Moving
(Selecting
Friday)
Time-Moving
(Selecting
Monday)
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0.90
1.00
Control (No Prime) Agency ("I"
Grammatical Subject)
Passivity ("ME"
Grammatical Object)
Condition
Proportion Selecting Farther Widget
Ego-Moving
(Selecting the
Distant Widget)
Time-Moving
(Selecting the
Closer Widget)