The generation effect and source memory [Elektronische Ressource] / vorgelegt von Sandra Daniela Mattern

Sujets

• Psychologie

Informations

• Publié par: universitat_mannheim
• Publié le: 01 janvier 2011
• Nombre de lectures: 12
• Langue: English
• Poids de l'ouvrage: 1 Mo

Extrait

UNIVERSITÄT MANNHEIM
The Generation Effect
and
Source Memory
Inauguraldissertation zur Erlangung des akademischen Grades
einer Doktorin der Sozialwissenschaften
der Universität Mannheim
vorgelegt von
Dipl.-Psych. Sandra Daniela MatternDekan der Fakultät für Sozialwissenschaften:
Prof. Dr. Berthold Rittberger
Gutachter: Prof. Dr. Edgar Erdfelder
Prof. Dr. Rüdiger Pohl
Tag der Disputation: 20. Dezember 2010“A curious peculiarity of our memory is that things are impressed better
by active than by passive repetition.”
(James, 1890, p. 646, italics added)
“Most of us have probably encountered the informally expressed sentiment
that there is an especial advantage to learning by doing,
or that some kind of active or effortful involvement of the person in the learning process
is more beneficial than merely passive reception of the same information.
To what extent does this general notion have solid empirical support,
as opposed to a casual or anecdotal base, [...]?”
(Slamecka & Graf, 1978, p. 592)Summary
Superior memory performance for self-generated (i.e., incomplete) items compared to read
(i.e., complete) items is called the positive generation effect, whereas the reverse pattern is
called the negative generation effect. For item memory tasks, a positive generation effect
typically occurs (cf., Slamecka & Graf, 1978). In contrast to this, no clear picture exists as to
whether a positive or a negative generation effect is bound to emerge for source memory of
perceptual attributes, due to empirical evidence for both outcomes. Therefore, the two lines
of research investigated in the course of the present dissertation deal with the generation
effect and source memory. They aim to shed light on present inconsistencies and
contradictions as well as to illuminate unanswered questions. All data were analysed using
the multinomial processing tree model for crossed source dimensions (Meiser & Bröder,
2002).
In the first line of research, I addressed the role of the processing of perceptual attributes
and of the processing of internal states in memory for the degree of completeness. Within a
generation effect paradigm, memory for source attributes of an item can be studied in
different ways. When considering the processing account by Mulligan (2004) and the dual-
hypothesis by Riefer, Chien, and Reimer (2007), contradictory predictions exist for the
source memory dimension degree of completeness – namely a negative generation effect in
the former and a positive generation effect in the latter case. To overcome these
contradictions, I hypothesised two processing modes: the processing of and memory for (a)
perceptual attributes (PA) and (b) internal states (IS). PA and IS processing were
implemented via weak instructions and strong instructions. Contrary to expectations, no
effect of instruction type could be found. In contrast, across types of instructions a null
effect emerged, when instructions were weak. When instructions were strong, however, a
positive generation effect occurred for both types of instruction. This latter finding is in line
with Riefer et al. (2007) and indicates the importance of self-reference (Rogers, Kuiper, &
Kirker, 1977) in source memory and in generation effect studies.
In the second line of research, I investigated the role of increased conceptual processing for
memory of presentation colour, for which a negative generation effect has been foundconsistently (e.g., Mulligan, 2004; Mulligan, Lozito, and Rosner, 2006; Riefer et al., 2007).
Some theories state that conceptual processing takes place to a higher degree for self-
generated items than for complete items (Donaldson & Bass, 1980; Graf, 1980; Mulligan,
2004; Slamecka & Graf, 1978), while other theories additionally claim that perceptual
processing takes place to a lesser degree for incomplete as compared to complete items
(Mulligan, 2004). These points were called into question by Mulligan et al. (2006). They
challenged the appropriateness of a processing account which emphasises perceptual and
conceptual processing, and considered it too specific. Instead, Mulligan and colleagues
suggested a more general processing account, which is concerned with visual and non-visual
processing instead. Since self-generation of items is strongly indicated to lead to higher
conceptual processing, I argued that whether or not an incomplete item has a schema-
typical colour influences colour memory performance; moreover, I theorised that it does so
to a higher degree for incomplete items than is true for simply read items. To test these
predictions, experiments were conducted, in which the supposed strength of the effect of
conceptual processing was varied. When less natural stimuli were employed, no effect of
schema-typicality of presentation colour or of completeness of the items could be found on
source memory for colour. In contrast, when more natural stimuli were employed, a
negative generation effect occurred. However, no additional effect of incompleteness and
thus no additional effect of the theorised increased conceptual processing of incomplete
items emerged. The present findings can thus be interpreted as evidence in favour of
Mulligan et al. (2006), who suggest a more general visual and non-visual processing account.
I conclude, (a) that self-reference plays a critical role for generation effect studies when
attempting to investigate source memory, and (b) that conceptual and perceptual processing
seem to play a less critical role than assumed – results could rather be interpreted as
evidence in favour of a more general account involving visual and non-visual processing.
In addition, crucial design issues for any research agenda concerned with the generation
effect were revised and evaluated for the current studies. Finally, I discussed remaining
research questions and suggested further research opportunities as well as future directions.Contents
1 Introduction 1
1.1 Item Memory and Effects of Self-Generation 3
1.2 Source Memory and Effects of Self-Generation 9
1.3 Interim Summary and Outlook 16
2 The role of the processing of perceptual attributes and internal states 17
2.1 Introduction to Chapter 2 17
2.2 Experiment 1 41
2.2.1 Methods 41
2.2.2 Results 48
2.3 Experiment 2 60
2.3.1 Methods 61
2.3.2 Results 62
2.4 Discussion of Chapter 2 68
3 The role of conceptual processing 77
3.1 Introduction to Chapter 3 77
3.2 Experiments 3, 4, and 5 95
3.2.1 Experiment 3: Methods 100
3.2.2 Experiment 3: Results 108
3.2.3 Experiment 4: Methods 118
3.2.4 Experiment 4: Results 122
3.2.5 Experiment 5: Methods 131
3.2.6 Experiment 5: Results 133
3.3 Discussion of Chapter 3 140
4 General Discussion 149
5 References 165
___________________________________________________________________________
vii6 Appendix 177
A Studies Implementing a Reality Monitoring Paradigm or an
External Source Monitoring Paradigm 179
B Stimuli Employed in Experiments 1 and 2 181
C Original Wordings for Test Options in Experiments 1 and 2 183
D Correct and Erroneous Responses Given During Study in Experiments 1 and 2 185
E Descriptions of Pilots I, II, IIIa, and IIIb 187
F Sample Descriptions of the Pilot Studies 189
G Stimuli Used in Experiment 3 191
H Stimuli Used in Experiments 4 and 5 193
I Naming Errors at Study in Experiments 3, 4, and 5 195
Ehrenwörtliche Erklärung
Danksagung / Acknowledgements
___________________________________________________________________________
viii