Transcript Lure Items - Middlebury College: Community Home Page

Does Distinctive Visual Information Enhance
monitoring of false memories?
2
Osborne ,
1
Stone ,
Abstract
 Materials: 96 sets of ten items (referred to as themes henceforth) from
Nelson, McEvoy, and Schreiber (1998)
 All items in a given theme produced a single item (the lure item) in free
association with some nonzero probability
 Theme items: rose, stem, blossom, lily, vase, dandelion, orchid, tulip,
petals, daisy
 Lure item: flower
 Unusualness of visual information at study was manipulated by presenting
study items in unusual-looking fonts (see Figure 3; Arndt & Reder, 2003;
Reder, Donavos, & Erickson, 2002).
 Previous research has indicated that the study of visual information
reduces semantic false recognition (Arndt & Reder, 2003; Schacter,
Israel, & Racine, 1999; Smith & Hunt, 1998).
Figure 2: Experiment 1 Results
Correct Recognition
 When unusual-looking fonts (e.g., Gadzoox) are uniquely associated
with study words, false recognition is lower than when unusuallooking fonts are associated with multiple study words (Arndt &
Reder, 2003).
1
Exp. 1: Correlated
Study
ROSE
STEM
BLOSSOM
LILY
VASE
DANDELION
ORCHID
TULIP
Study
ROSE
STEM
BLOSSOM
LILY
VASE
DANDELION
ORCHID
TULIP
A'
0.6
 Materials: 48 themes from Nelson, et al. (1998). For each theme, the four
strongest associates (mean associative strength = .585) and four weaker
associates (associative strength between .02 and .10; M = .063) were
selected to serve as study items.
 Strong associates: trout, tuna, salmon, bait
 Weak associates: line, lobster, lake, mussel
Figure 4: Experiment 2 Results
Correct Recognition
0.3
0.2
0.1
False Recognition
1
0.9
Correct Recognition
0.8
0.6
Unique
0.5
0.4
Speeded
Unspeeded
Retrieval Speed
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
Experiment 1 Conclusions
 Encoding-based explanation insufficient.
Match
Mismatch
 Editing-based monitoring processes insufficient.
 Results suggest possibility that employment of monitoring processes can inflate lure
errors.
 One mechanism by which this could occur is if lure representations develop
associations with encoded features of studied items.
 Activation-monitoring theory (Roediger, et al., 2001) suggests this tendency
varies with activation of lure item at encoding.
Unique
Correlated
p(false alarms)
0.3
0
 More errors when lures shown in a font used to present
their studied associates (Arndt & Carney, 2004).
 More errors when lures shown in a font used to present
their strong associates (Experiment 2).
 The association between the visual format of studied items
and unstudied lure items may lead to impaired monitoring
processes.
 Higher errors in the correlated condition than the unique
condition emerged late in retrieval (Experiment 1).
 This hypothesis is consistent with many of the features of
false recognition in the DRM paradigm.
 Recollection phenomenology associated with lure items
(Roediger, & McDermott, 1995).
 Warnings do not eliminate lure errors (Multhaup &
Conner, 2002).
 Source monitoring produces higher error rates than oldnew recognition (Hicks & Marsh, 2001).
Arndt, J., & Carney, J. (2004). Distinctive information and false
recognition: The role of encoding and retrieval factors. Poster
presented at the Annual Meeting of the Psychonomic Society,
Minneapolis, MN.
Hicks, J.L., & Marsh, R. (2001). False recognition occurs more
frequently during source identification than during old-new
recognition. Journal of Experimental Psychology: Learning,
Memory, and Cognition, 27, 375-383.
Johnson, M.K., Hashtroudi, S., & Lindsay, D.S. (1993). Source
monitoring. Psychological Bulletin, 114, 3-28.
Multhaup, K. S., & Conner, C. A. (2002). The effects of considering
non-list sources on the Deese-Roediger-McDermott memory
illusion. Journal of Memory and Language, 47, 214-228.
Lure False Alarms
Correlated
0.1
 Results from these studies and those reported by Arndt and
Carney (2004; see also Arndt, in press) are consistent with
the view that lure representations develop associations with
visual features of studied items.
Arndt, J., & Reder, L.M. (2003). The effect of distinctive visual
information on false recognition. Journal of Memory and Language,
48, 1-15.
High
Low
Associative Strength
0.2
False Recognition
p(false alarms)
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0.7
Lure A'
p(hits)
Figure 1: Arndt & Carney (2004) Results
Conclusions
References
0
Correlated
Strong: FISH
Weak: FISH
Experiment 2
Correlated
Fast
Slow
Retrieval Speed
1
Lee
Lure Items
Unique
0.5
 Retrieval-based models (e.g., Johnson, Hashtroudi, & Lindsay, 1993)
suggest that monitoring is greater in the unique condition than in the
correlated condition.
Unique
Strong: Trout
Weak: Line
0.4
Retrieval-Based Models
Mismatch
Trout
Tuna
Salmon
Bait
Line
Lobster
Bait
Mussel
Study Items
 At test, lure items were presented in the font used to present their strong
associates or weak associates (see Figure 3).
0.7
 Activation of lure representations is lower when unusual-looking
fonts are studied (Roediger, et al., 2001).
Match
Study
 Strong associates were presented in one font,while weak associates were
presented in a second font.
0.8
 Presentation of study items in unusual-looking fonts enhances
encoding of item-specific information and reduces encoding of
relational information.
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
Exp. 2 Study List Exp. 2 Test Conditions
 Lure item: fish
0.9
 False recognition has been shown to be higher when a font presented
during encoding is reinstated at retrieval (Arndt & Carney, 2004; see
Figure 1).
Encoding-Based Models
Exp. 1: Unique
 Design: Font condition (correlated vs. unique) and Retrieval Speed (speeded
vs. unspeeded) both varied within subjects
Prior Research
& Karen
Figure 3: Experiment 1 & 2 Conditions
Experiment 1
Two experiments evaluated the role of encoding-based and retrievalbased factors in the production of false recognition in the DRM
paradigm. The first experiment varied the association of unusuallooking fonts with DRM themes (correlated or not correlated) and the
duration of time allotted for subjects to utilize retrieval monitoring. The
second experiment manipulated whether the font in which an
unstudied lure item was presented at test was used to present strong
or weak associates of the lure, as measured by Backward Associative
Strength (BAS). The combined results of these experiments suggest
that subjects utilize visual format as a retrieval cue in an effort to
distinguish between studied and unstudied items, but that this
monitoring strategy can be harmful when lure activation during
encoding leads to the development of associations between fonts
used to present study items and lure item representations.
1
Carney
Ashley
Meghan
John
1Middlebury College & 2Duke University
p(hits)
Jason
1
Arndt ,
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
Nelson, D.L., McEvoy, C.L., & Schreiber, T.A. (1998). The
University of South Florida word association, rhyme, and word
fragment norms. http://w3.usf.edu/FreeAssociation/
Roediger, H. L., III., & McDermott, K. B. (1995). Creating false
memories: Remembering words not presented in lists. Journal of
Experimental Psychology: Learning, Memory, and Cognition, 21,
803-814.
Roediger, H.L., III, Watson, J.M., McDermott, K.B., & Gallo, D.A.
(2001). Factors that determine false recall: A multiple regression
analysis. Psychonomic Bulletin & Review, 8, 385-407.
Schacter, D.L., Israel, L., & Racine, C. (1999). Suppressing false
recognition in younger and older adults: The distinctiveness
heuristic. Journal of Memory and Language, 40, 1-24.
Smith, R., & Hunt, R.R. (1998). Presentation modality affects false
recognition. Psychonomic Bulletin & Review, 5, 710-715.
High
Low
Associative Strength
Experiment 2 Conclusions
 Evidence that lure representations develop associations with presentation
features of study items.
 Activation-monitoring theory (Roediger, et al., 2001) suggests this
tendency varies with activation of lure item at encoding.
Please address correspondence to
Jason Arndt ([email protected])