ArticlePDF Available

Effects on recall of signals to text organization



College subjects read and recalled one of two versions of the same text. One version of the text contained signals to both the general organization of the text and to the organization of specific sentences in the text; the alternate version did not contain organizational signals. Overall recall was better for the text with signals than for the text without signals because recall of specifically signaled information was aided by signaling. The magnitude of the signaling effect was related to subjects’ text recall abilities: Signaling effects were larger for better recallers than for poorer recallers.
the Psychonomic Society
1985. 23 (4). 374-376
Effects on recall of signals to text organization
Kentucky. Lexington. Kentucky
(Susan M. Belmore, Sponsor)
College subjects
and recalled one of two versions of
same text. One version of
contained signals to both
general organization of
organization of specific
sentences in
version did not contain organizational signals. Overall recall
text without signals because recall of specifi-
cally signaled information was aided by signaling. The magnitude of
signaling effect was
to subjects'
recall abilities: Signaling effects were larger for
for poorer recallers.
In order to comprehend a text, a reader must represent
the structure
the text as well as its content (Kintsch &
van Dijk, 1978). In processing a text's structure, a reader
must recognize both the relative importance
text propo-
sitions and the organization of those propositions. A writer
can facilitate the reader's task by using a variety
nals to indicate a text's structure and to emphasize par-
ticular information (Meyer, 1975; van Dijk, 1979).
A text signal is a writing device that does not add new
content to a text; rather it emphasizes certain aspects of
a text (Meyer, 1975). There are many different kinds of
signaling devices. Some signals emphasize specific text
content (e.g .,
is important to stress
, whereas
other signals emphasize the organization of information
in a text (e.g., numbering). Signals may operate at high
levels (e.g. , titles) or low levels (e.g., underlining) of text
structure. Finally, signals may involve typographical cues
which distinguish them from the text per se (e.g., head-
ings), or they may be embedded in the text (e.g., preview
sentences). The present investigation is concerned with
the combined effects of two text-embedded signaling
devices which function to emphasize text organization.
The issue
primary concern is how the organizational
signals affect text recall. The results of the experiment
also .allow consideration of the nature of individual differ-
ences in the effects of organizational signals.
is well known that organization is an important de-
terminant of recall of word lists (e.g ., Tulving &Pearl-
stone, 1966) and
text (e.g., Frase , 1969). Thus, sig-
naling the organization of a text should facilitate its
subsequent recall by making that organization more ex-
plicit and salient to the reader. One way to emphasize text
This research was supported by a University of
Grantand bya UKRF
. I
like to thank DavidA.
and Elizabeth Pugzles Lorch for their careful critiques of earlier ver-
sions of this paper.
for reprints shouldbe addressed to: R. F.
Lorch, Jr., Department of
tucky, Lexington, KY 40506-0044.
organization is with signals that are physically distinct
from the text, such as headings (e.g. ,
is evidence that headings selectively aid recall
tant text information (Brooks, Dansereau, Spurlin, &Hol-
ley, 1983; Loman &Mayer, 1983). Another way to sig-
nal organization is with preview sentences, logical
connectives, and other devices embedded in the text. In
contrast to findings on the effects of headings, studies of
text-embedded organizational signals have found little evi-
dence of effectiveness . Several studies have failed to
demonstrate that signaling text structure facilitates over-
all recall (Britton , Glynn, Meyer , &Penland, 1982; Ir-
win, 1982; Meyer, 1975). Those studies reporting effects
found better recall only for signaled information and only
for underachieving or poor readers (Marshall &Glock,
1978; Meyer, Brandt, &Bluth, 1980).
Previous failures to demonstrate effects of text-
embedded organizational signals may be due to the use
of overall recall as the dependent measure. When other
types of text signals have been studied, the presence
signals has often been found to facilitate recall of signaled
information while inhibiting recall of unsignaled infor-
mation (Fowler &Barker, 1974; Glynn &DiVesta, 1979;
Loman & Mayer, 1983). The current experiment thus ex-
amined the effects of text-embedded organizational sig-
nals on signaled and unsignaled information. Subjects read
and free recalled a text that either did or did not include
signals. The signaled version of the text included both
preview sentences indicating the major sections of the text
us consider the first piece of evidence.
numbering signals indicating the organization of specific
facts presented in the text (e.g.,
First, .
jects' overall recalls were broken down into two compo-
nents. First , recall was scored for content identified as
being specifically indicated by numbering signals
Second, recall was assessed for the remaining text
the organizational signals
affect recall selectively, then specific recall scores should
be higher for the signaled text, and nonspecific recall
scores should be higher for the unsignaled text.
Copyright 1985 Psychonomic Society, Inc. 374
Asecondary concern of the current study is whether
readers of different text-recall abilities respond differently
to text signals . Two competing hypotheses may be pro-
posed to explain how signaling effects might vary with
differences in text-recall ability. One possibility is that
good recallers do not need the help of text signals to
process a text's organization, whereas poor recallers are
deficient in organizational skills and, thus, benefit from
devices that make a text's organization more salient
(Meyer et al. , 1980). This hypothesis predicts that sig-
nals will aid poor recallers more than good recallers. The
second hypothesis is that differences in sensitivity to or-
ganizational cues are an important component of differ-
ences in text-recall ability (Brown, 1980). Good recallers
are good in part because they attend to organizational sig-
nals as they read . This hypothesis predicts that good
recallers will benefit more from signals than poor
recallers. In order to examine individual differences in
signaling effects , a measure of text recall performance
will be correlated with a measure
the effects of signal-
arelationship is demonstrated, the nature
relation will be observed in order to determine whether
signals are more beneficial to good or poor text recallers.
Subjects were 113 introductory psychology student s at the U niver-
sity of Kentucky who received credit for participation in the experiment.
The stimulu s text was a story about the kidnapping of Charles Lind-
bergh's baby (Wallechinsky &Wallace , 1978). In addition, the Nelson-
Denny (Brown, Nelson, &Denny, 1973) vocabulary test
item, mul-
tiple cho ice) was used in a distract or task .
Two versions of the text were written . Each ver sion was typed single
spaced on two pages. The signaled version contained preview sentences
at the start of mo st paragraphs. These sentences clearly indicated the
topic of the subsequent section of the text without contributing new in-
formation to their respective paragraphs . In addition, there were two
places in the text that involved a listing of evidence relevant to the kid-
napping case. The organization of this evidence was signaled by the
use of an introductory sentence and by numbering of each successive
piece of evidence . The unsignaled ve rsion of the text differed from the
signaled ve rsion in that the pre view sentences were deleted . Also, the
numbering s ignals were deleted and the initial parts of the correspond-
ing sentences were rewritten where necessary to maintaingrammaticality.
Subjects were run in groups of between 5 and 40 subjects in sessions
lasting approximately 45 min . Subject s were assigned at random to the
two experimental conditions (56 in the signal condition ; 57 in the no
signal condit ion) . They were instructed to read the text once at their
own pace in preparation for a subsequent recall test. When all subjects
had read the story , the texts were collected and the vocabulary test was
administered for 10 min. Finally , subjects gave a writt en free recall of
the story. The recall was self paced .
The unsignaled version of the text was analyzed into 43 idea units,
one idea unit for each major sentence clause in the text. Each subject's
recall protocol was scored with respect to the number o f idea units cor -
rectly reproduced from the text. Overall recall was broken down into:
(I ) recall of the 9 idea units that were specifically marked by number -
ing in the text (spe cific recall score) ; and (2) recall of the remaining
34 idea units (nonspecific recatl score). The scoring was done by ajudge
who was blind to the purpose s of the experiment . The j udge was trained
in the scor ing procedure until a satisfactory level of agreement was
achieved between the judge and author (kappa >.8).
Two sets of analyses were performed. The level
nificance exceeds .05 in all tests unless indicated other-
wise. The first set
analyses compared recall for the sig-
nal and no signal conditions. More idea units were recalled
for the text with signals (49.16% of the ideas) than for
the text without signals (44. 88 %) [t(111)=2.00]. Recall
of information associated with number signals was higher
for the text with signals (56.55 %) than for the text without
signals (45 .81 %). This effect was reliable when tested
over subjects [t(111) = 3. 13] and when tested over items
.53] . Nonspecific recall was not reliably differ-
ent for the signal (47.22%) and no signal conditions
, P >.1].
The second analysis concerned individual differences
in signaling effects. In this analysis , the specific recall
scores were interpreted as a measure of the signaling ef-
fect because only those scores were affected by the sig-
naling manipulation. The nonspecific recall scores were
interpreted as a measure
subjects ' text-recall abilities
because there was no effect of the signaling manipulation
on those scores. For each experimental condition, sub-
jects' specific recall scores were regressed on their non-
specific recall scores . The slope coefficient was greater
in the signal condition (1.157) than in the no signal con-
dition (0.615) [t(109)=2.51]. Looking at it another way,
when subjects within each condition were divided in half
according to their nonspecific recall scores , the differ-
ence in specific recall scores between the signal and no
signal conditions was greater for the good recallers
(15.5% difference) than for the poor recallers (3 .9%
The results of the experiment are easily summarized . First , the presence
of organizational signals selectively aided recall of those ideas which
were specifically marked by numbering signals. Further, better reca1lers
benefitted more than poorer recaller s from the presence of text signals.
Let us consider the implications of these findings.
This is the first demonstation of an effect of text-embedded organiza -
tional signals on overall recall when results are averaged across all sub-
jects . However , the effect on overall recall was primarily due to the
fact that signaling effects were greater for ideas that were specifically
associated with text signals than for ideas that were not specifically
marked. This result of a selective effect of signaling on recall is consis-
tent with several previous findings (Fowler &Barker, 1974; Glynn &
DiVesta, 1979; Loman &Mayer, 1983; Marshall &Glock , 1978).
Together, these results suggest that previous investigators failed to ob-
serve effects of organizational signals because they examined only overall
recall (Britton et al . , 1982; Meyer , 1975). Specific recall effects of the
organizational signals may have been obscured by this procedure.
Given that signals facilitate recall of the content they mark , we might
ask how the signals operate. Although the data do not resolve this is-
sue, several alternative mechanisms may contribute to the signaling ef-
fect. First, signals may cause reader s to pay more attention to the as-
sociated content with the consequence of better memory for the
information. This hypothesis might be tested by measuring reading times
for signaled and unsignaled sentences or by use of a secondary probe
reaction time task. Second , explicitly marking the organization
content may lead readers to represent that organization more completely .
Arepresentation of the organization of specific text content should al-
low for more systematic and complete recall of the specific informa-
tion. Third , signals may influence the order in which text information
is recalled . Signaled information may be produced earlier in recall than
unsignaled information with the consequence that signaled content is
subjected to less output interference than unsignaled content. Finally ,
signals may distinguish the associated content from unsignaled content
and confer greater importance on the signaled information. To the ex-
tent that the process of free recall entails some degree of editing of redun-
dant and unimportant information, the effect of signals may be to make
the associated information less likely to be deleted from recall.
The second finding of the experiment was that the effect of the or-
ganizational signals depended upon the text-recall ability of the subject.
One implication of this result is that previous investigations may have
failed to find effects of organizational signals in part because they aver -
aged results across subjects of differing abilities. The more important
implications of the finding concern the cognitive basis for the relationship.
Two competing hypotheses were offered concerning the relation be-
tween text-recall ability and the effects of signaling. First , if good and
poor recallers differ in how actively they employ organizational strate-
gies during reading, then including explicit signals to text organization
should have benefitted the passive, poorer recallers more than the good
recallers. Alternatively, good recal1ersmay construct more coherent text
representations because they employ available information to better or-
ganize their memories . The results supported the second hypothesis :
Subjects who were better recallers benefitted more from the organiza-
tional signals than poorer recallers . This result is similar to a previous
finding that underlining affected the recalls of slow learners less than
the recalls of medium or fast learners (Crouse &Idstein, 1972).
Final1y,the preliminary nature of these findings
be noted. First, the findings are
upon one text. They must be repli-
cated with different texts before they can be generalized. Second , it is
tempting to attribute the observed selective recall effects specifically
to the number signals rather than to the combined effects of the number
signals and preview sentences . However, the design of the experiment
does not allow such a conclusion . Future research should manipulate
the two types of signals independently . Third , the relat ion between sig-
naling effects and recall ability must be tested using a measure of recall
ability which is defined a priori and independently of the experimental
manipulations. Finally, other types of organizational signals must be
systematically investigated . This is a particularly important issue given
the finding that some organizational signals appear to benefit underachiev-
ing readers more than good readers (Meyer , 1980), in apparent
contrast to the findings of the current study .
, B. K.,
S. M.,
, B. J. F., '"
J. (1982). Effects of text structure on use of cognitive capacity dur-
ing reading.
Educational Psychology, 74, 51-61.
,L. W. ,
, D . F. ,
, J. E., '"
, C.
D. (1983). Effects
headings on text proce ssing.
tional Psychology, 75, 292-302 .
, A. L. (1980). Metacognitive development and reading. In R. J.
Spiro , B. C. Bruce , &W. F . Brewer (Eds.), Theoretical issues in
reading comprehension (pp. 453-481 ). Hillsdale , NJ: Erlbaum .
J. I. , N
, M.
, E. C. (1973) .
Denny reading test. Boston: Houghton Mifflin .
, J . H. , '"
P. (1972) . Effects of encoding cues on prose
learning .
Educational Psychology, 63, 309-313 .
, R. L. , '"
A. S . (1974). Effectiveness othighlighting
for retention of text material . Journal
Applied Psychology, 59,
L. T. (1969). Paragraph organization of written materials : The
influence of conceptual clustering upon the level and organization of
Educational Psychology, 68, 394-401.
, S. M ., '"
F. J. (1979). Control of prose processing
via instructional and typographical cues .
chology, 71, 595-603 .
, J. W. (1982). The effects of coherence explicitness on college
readers' prose comprehension.
Reading Behavior, 14,
275-284 .
, W., '"
T. A. (1978). Toward a model of discourse
comprehension and production.Psychological Review, 85, 363-394 .
, N. L., '" MA
R. E . (1983) . Signaling techniques that in-
crease the understandability of expository prose .
tional Psychology, 75, 402-412 .
, N., '"
M. D. (1978). Comprehension of connected
discourse: A study into the relationship between the structure of text
and informat ion recalled . Reading Research Quarterly, 14, 10-56.
, B. J. F. (1975) .
prose and its effects in
memory. Amsterdam : North-Holland.
, B. J . F. ,
, D . M. , '"
, G. J. (1980). Use of top-
level structure in text: Key for reading comprehension of ninth-grade
students. Reading Research Quanerly, 16, 72-102 .
E., '"
Z. (1966). Availability versus accessi-
bility of information in memory for words.
ing &Verbal Behavior,S, 281-291.
DUK, T . A. (1979). Relevance assignment in discourse compre-
hension. Discourse Processes, 2, 113-126.
D., '"
I. (1978).
people's almanac #2.
New York: Bantam Books.
(Manuscript received for publication April
... 1978;Fishman, 1978;Keenan, Baillet, & Brown. 1984;Kieras, 1978;Kintsch & Yarborough, 1982;Lorch & Lorch, 1985;Mayer, 1983;Meyer, Brandt. & Bluth, 1980;Meyer & Freedle. ...
... option com- mitte es have gone so far as to neglect to consider learnability explicitly at all. using instead such proxies as readability 6Hartley. Bartlett. & Branthwaite. 1980;Hartley. Goldie. & Steen. 1979;Lorch & Chen. 1986;McLaughlin-Cook. 1981;Spyrldakis & Standal, 1986. 7Bransford & Johnson. 1972Dooling & Lachman. 1971;Glynn & DiVesta. 1979;Lorch. 1985;Parker. 1962;SchaUert. 1976. formulas and copyright date (Tyson-Bernstein, 1988). Per- haps they lack confidence in their judgments of learnability. This raises the important question of whether people can make accuratejudgments about the learnability oftextbooks. ...
review the evidence for two empirically based conclusions about the impact of text quality on learning [in undergraduates] / describes how a theory of learning from text can be applied to rewriting text / projects future practice and research in text quality quality of writing and learning: the empirical evidence / quality of writing and judgments of learnability: an experiment with 20 texts / a cognitive theory of learning from text applied to rewriting text to improve learning / an experiment in rewriting to improve learning / implications for textbook selectors, publishers, and researchers (PsycINFO Database Record (c) 2012 APA, all rights reserved)
... b) Aquellas que, también bajo la forma de resumen o sumario, se ofrecen al final del texto a modo de conclusión. c) Señalizaciones, llamadas ordinales o numerales por algunos autores (Lorch, 1985;Lorch y Chen, 1986), suelen utilizarse cuando se presenta un argumento que expresa diversos puntos de vista o diversas partes. En este caso, cada uno de ellos suele ir precedido por un número, letra o palabra «e.g. ...
... Al igual que ellos, todos ofrecen conjuntamente una panorámica más amplia y clara de la macroestructura del material escrito (Lorch y Chen, 1986). Este efecto selectivo de la información recordada ha sido puesto de manifiesto, tanto en el uso de señalizaciones aisladas, ya sean éstas numéricas (Lorch, 1985;Lorch y Chen, 1986), conectivas lógicas (Irwin y Pulver, 1984) o títulos (Wilhite, 1986), como en la combinación de varias, tales como frases previas, conectivas y otros recursos (Loman y Mayer, 1983: Mayer, Cook y Dyck, 1984Meyer, 1984;Meyer, Brandt y Bluth, 1980;Meyer y Rice, 1982;León, 1989;León y Carretero, en prensa). ...
Full-text available
El presente trabajo analiza, desde un enfoque interactivo, algunas características del texto y del lector. Desde esta perspectiva se revisan algunos de los modelos de intervención encaminados a facilitar en el sujeto el producto final de la comprensión lectora. Como es sabido, esta intervención puede introducirse tanto desde el mismo material escrito, a través de técnicas y recursos sobre planes de escritura que aplica el escritor, como desde las características del lector, como es el conocimiento previo y el uso de estrategias lectoras más eficaces en los sujetos. Desde esta última perspectiva incidimos especialmente en aquellos entrenamientos de estrategias encaminadas a potenciar un conocimiento y aplicación de la estructura organizativa de los textos. This paper provides an assessment of the interactive approach between text and reader features. Within this framework several intervention models designed to facilitate the subject's attainment of reading comprehensión are reviewed. As is usually known, intervention procedures may be based either on the written material itself, by means of the writer's own techniques and resources, or on the use and modification of more effective reading strategies on the subject's part. From this latter perspective, training in strategies intended to foster the knowledge and application of the text organizational structure is emphasized.
Full-text available
Purpose Since researchers have utilized text signals to develop a mass of within-document visualization analysis tools for reading aid in a long document, there is an increasing need to study the relationship between readers’ behavior of using text signals for navigation and their reading performance in the tools. The purpose of this paper is to combine the text signals using behavior and reading performance in two kinds of analysis tools to verify their relationship and discover whether there is any efficient reading strategy when using text signals to navigate a long document. Design/methodology/approach The methodology is a case study. The authors reviewed related literature first. After explaining the design ideas, interface and functions of THC-DAT and BOOKMARK, which are two reading tools utilizing two main kinds of text signals, one utilizing topics and the other utilizing headings for reading aid, a case study was presented to collect click data on the text signals of participants and their reading effectiveness (score) and efficiency (time). Findings The results confirm that the text signals using behavior for navigation has a significant impact on reading efficiency and no impact on reading effectiveness in both BOOKMARK and THC-DAT. The discrete degree of clicks behavior on text signals has an impact on reading efficiency. The using behavior of different types of text signals has different impacts on reading efficiency. Research limitations/implications Using text signals for navigation time evenly can help improve reading efficiency. And a basic strategy suggested to readers is focusing on reducing their time to find answers when using text signals for navigation in a long document. As to utilizing the two different kinds of text signals, readers can have different strategies. Accordingly, personalized recommendation based on interval of adjacent clicks will help to improve computer-aided reading tools. Originality/value This paper combines the text signals using behavior for navigation and reading performance in two kinds of visual analysis tools, studied the relationship between them and discovers some efficient reading strategies when using text signals for navigation to read a long document.
A primary goal of reading is to understand the writer’s message correctly. To accomplish this goal, a reader must identify the information in the text that is most relevant to the author’s message. A writer can facilitate the reader’s task by using a variety of devices to signal relevant information. In the present study, we investigated the effects of signals on text comprehension and recall in different groups. Subjects were 48 Experts (post-graduate students) and 96 Novices (48 good readers and 48 poor readers, junior high school students). All subjects read two expository texts at different levels of difficulty. There were two versions for each text: with signals and without signals. Two recall tests were run: immediate free recall and delayed free recall. The results demonstrated that signals directed attention to the macrostructure they marked and led to a better encoding of the organization of target information in all groups. This effect increased with the difficulty of the text and the skill of the reader.
The combined effect of number signals and preview sentences on readers’ recall of text material was examined. Subjects who read a passage in which both types of signals were used recalled significantly more signal-relevant information than did subjects in a number-signals only, a preview-sentence only, or a control condition. The effectiveness of number signals is discussed in terms of both encoding and retrieval effects. The influence of preview sentences is discussed in terms of how these devices may guide readers’ attention.
Do experienced readers alter their processing of topics in a text according to task demands? Subjects read a text in preparation for a task that either emphasized topic information (outlining) or did not emphasize topic information (verification). Reading times were recorded for sentences which introduced either major or minor shifts of text topics. Subjects read topic sentences more slowly if they introduced major topic shifts, but the magnitude of the effect depended upon the task and the subjects’ text recall ability. The effect was greater for better recallers than for poorer recallers. Also, the effect was greater in the outline task than in the verification task, but only for better recallers. The results indicate that: (a) Readers determine topic interrelations as they encounter new topics during reading; (b) better recallers are more consistent about inferring topic interrelations; and (c) better recallers are more flexible in their processing of topic information.
How do readers respond to signals in a text that certain information is relevant? Subjects in Experiment 1 read texts a sentence at a time while their reading times were recorded for specific target sentences. Subjects took longer to read a summary sentence if the preceding sentence signaled it as a summary than if no preceding signal was provided. Further, the effect of signaling a summary was larger for poor readers than for good readers. There was no corresponding effect of signaling on reading times for sentences marked as important statements, but Experiment 2 demonstrated that recall of important text information was enhanced if the information was signaled rather than unsignaled. The results demonstrate that experienced readers use text signals to guide their attention to relevant information in a text.
This paper reports on a study of 354 pupils in the top form of Dutch primary schools spread over 18 classes and two research conditions (an experimental and a control condition). The experimental condition involved teaching the recognition of two text structures, the classification structure and the causation structure. Pupils were also taught how to make schematics in which the main points of the text are arranged in line with the text structure. The study shows that pupils of around 12 years old are able to recognize the text structures involved and can learn to master a complex study strategy like 'making schematics'. These are the main results of the experimental lesson series with which pupils in the experimental condition were taught. The study also revealed transfer-effects. The experimental lesson series proved to have an effect on the ability to infer the main idea of a text and spontaneous application of the study strategy learnt.
Signals are writing devices that emphasize aspects of a text's content or structure without adding to the content of the the text. Findings are reviewed for studies of several different types of signaling devices, including: titles, headings, previews, overviews, summaries, typographical cues, recall sentences, number signals, importance indicators, and summary indicators. Most investigations have examined how the presence of signals in a text affects subsequent memory for the text. Virtually all types of signals produce better memory for information they cue in a text, whereas memory for unsignaled information often is unaffected. Less attention has been directed to signaling effects on other cognitive processes, such as attention, basic reading processes, and comprehension. It is argued that an understanding of how signals influence these processes will contribute to the application of signaling research to reading and writing instruction and to our general understanding of reading.
Full-text available
The aim of this study was to gain insight into the extent to which health education text writers apply writing principles derived from cognitive psychological theory. Seventeen professional text writers of health education materials participated in a qualitative study, consisting of a rewriting task combined with a think-aloud procedure and a semistructured interview. The verbal data were explored carefully in light of seven text coherence principles that have proven effective in cognitive psychological research to increase text comprehension. Findings indicate text writers vary in their ideas and use of coherence principles to make a text comprehensible. It is argued that the health education profession can benefit greatly from knowledge about cognitive text processing and cognitive coherence principles for realizing effective comprehension of written health education messages.
BONNIE J.F. MEYER DAVID M. BRANDT GEORGE J. BLUTH Arizona State University THE STUDY INVESTIGATEnSi nth-grade students' use of a reading strategy (the structure strategy) which focuses on following the organizational structure of text in order to determine what is important to remember. Texts read were well organized with problem/solution or comparison structures; signaling varied the saliency of these structures. Signaling effects were expected to interact with mastery of the structure strategy. Regardless of signaling, good comprehenders on the Stanford Achievement Test were expected to follow the structure strategy while poor comprehenders were not. However, comprehension underachievers (vocabulary substantially above comprehension test scores) were expected to follow the structure strategy only when signaling was present. Most predictions were supported; the structure strategy appeared to be a particularly effective retrieval mneumonic. Its development with age across different discourse types is discussed. Also characterized is the approach to reading and retelling of ninthgrade students who do not employ this structure strategy.
This study was designed to discover how certain aspects of text affected comprehension. To do this, 4 aspects of text structure were manipulated: 1) If-then relations were either explicitly stated or implied in the text; 2) adjectives were given either in the comparative/superlative form or in their simple form in the text; 3) the main idea was placed either at the beginning or the end of the text; 4) a designated clause was placed either at the beginning or the end of a designated sentence in the text. These manipulations were found to affect the recalls of community college subjects but not those of Ivy League college subjects. The data were used to support Frederiksen's (1975b) model of the structure of text and of memory and to define 2 different populations of readers./// [French] Cette etude a ete concue dans le but de découvrir comment certains aspects d'un texte agissent sur la compréhension. Afin de découvrir ceci, 4 aspects de la structure du texte ont été manipulés: 1) les rapports causaux ont été précisés s'ils ne découlaient pas implicitement due texte; 2) dans le texte les adjectifs ont été présentés ou bien à leur forme comparative/superlative ou bien à leur forme simple; 3) l'idée principale a été située ou au commencement ou à la fin du texte; 4) une proposition spécifique a été située ou au commencement ou à la fin d'une phrase indiquée dans le texte. On a remarqué que ces manipulations ont agi sur le rappel du texte chez des étudiants qui fréquentent des universités qui se spécialisent dans l'éducation du grand public (community colleges). On a noté aussi que ces manipulations n'ont eu aucun effet sur des étudiants qui fréquentent des universités privèes d'élites (Ivy League colleges). Les données expérimentales ont été utilisées pour soutenir le modéle de Fredericksen (1975b) de la structure du texte et de la mémoire et comme moyen de délimiter 2 groupes distincts de lecteurs./// [Spanish] Este estudio fue disenado para descubrir cómo ciertos aspectos de un texto afectan su comprensión. Para ello, se operó con 4 aspectos de la estructura de un texto: 1) relaciones del tip si-entonces contenidas en el texto en forma explícita o tácita; 2) los adjetivos fueron usados en su forma simple, comparativa o superlativa; 3) la idea principal fue puesta al comienzo o al final del texto; 4) una frase determinada fue colocada al comienzo o al final de una oración en el texto. Se observó que estas operaciones afectaron la memorización entre individuos de "colleges" comunitarios pero no afectó a individuos de "colleges" Ivy League. Se usaron los datos para respaldar el modelo de Frederiksen (1975b) de estructura de texto y de memorización y para definir 2 diferentes grupos de lectores.
This study was designed to investigate the comparative effects of the ex-plicitness of two types of intersentential coherence relations on college students' comprehension. The explicitness of argument repetitions (AR) and connective concepts (CC) was varied across forms of an experimental passage resulting in the following four versions: explicit AR/implicit CC, explicit CC/implicit AR, both explicit, both implicit. College students were randomly assigned to one of the four versions. They silently read the passage and then immediately wrote their recall protocols. The results indicated that there were no significant differences among the groups at either the micro or macro levels in terms of reading times, amount recalled, or reading time/amount recalled ratios. There were, however, significant differences among the variances in the four groups. These results are discussed in terms of processing theories and in terms of possible individual difference variables.
An informal analysis is given of the notion of “relevance” as used in the description of discourse and discourse comprehension. The analysis is restricted to various kinds of semantic relevance. A distinction is made between relevance at the level of sentences, as it is usually discussed in terms of the opposition of topic and comment, and relevance at the global level of whole texts. In the latter case, relevance is identical with the theme or gist of a text, which is made explicit in terms of macrostructures. At both levels a further distinction is made between “normal” relevance and contrastive or differential relevance. These forms of textual relevance are distinguished from contextual relevance, which is assigned to properties of the text on the basis of the cognitive set (actual knowledge, beliefs, opinions, wishes, attitudes, or tasks) of the reader. Finally, the problem of the cognitive processes and representations involved in these various kinds of relevance assignments is discussed.