Örebro Universitet
Institution för beteende-, social och rättvetenskap Psykologi
THE NOVELTY EFFECT IN MONOLINGUAL AND BILINGUAL CHILDREN
Psykologi C VT 2006 Författare: Nestor Vinas Guasch Handledare: Reza Kormi-Nouri
NOVELTY EFFEKTEN HOS ENSPRÅKIGA OCH TVÅSPRÅKIGA BARN1
Nestor Viñas Guasch
Institution för beteende-, social och rättvetenskap Psykologi , Örebro Universitet
Syftet med denna studie är att undersöka minnesprestationer hos enspråkiga- och tvåspråkiga barn och om ny information känns igen på ett annat sätt än bekant information hos enspråkiga jämfört med tvåspråkiga barn samt hos killar jämfört med tjejer. Ett stickprov på 49 elever från svenska skolor (28 tvåspråkiga och 21 enspråkiga, 24 tjejer och 25 killar) användes i studien. Både enspråkiga- och tvåspråkiga fick i uppgift att komma ihåg och besluta om olika ordlistor. De fick också två test om semantiskt minne. Efteråt fick de bedöma sin förmåga i svenska språket (skriva, läsa, tala och förstå) och tvåspråkiga bedömde även sin förmåga i hemspråket. Inga skillnader fanns mellan grupperna avseende resultaten av minnsestesterna. Vidare upptäcktes inga skillnader i att komma ihåg ny och bekant information hos någon av grupperna.
Nyckelord: Novelty effekten, minne, tvåspråkighet, språkintegration. 1 Psykologi C, VT 2006. Handledare: Reza Kormi-Nouri.
THE NOVELTY EFFECT IN MONOLINGUAL AND BILINGUAL CHILDREN 1
Nestor Viñas Guasch
Department of behavioral-, social- and legal sciences Psychology , Örebro University
This study aims to examine memory performance in monolingual and bilingual children, and whether novel information (compared to familiar information) is differently recognized by bilingual children compared to monolingual children, and for boys compared to girls. A sample of 49 pupils from Swedish schools (28 bilingual children and 21 monolingual children; and 24 girls and 25 boys) was used in this study. Both bilinguals and monolingual had to remember or make decisions about different lists of words. They also performed two tasks of semantic memory. After that, they had to rate themselves in their proficiency of Swedish knowledge (writing, reading, speaking and understanding) and knowledge in mother tongue (only for bilinguals). In all memory tasks, no differences were found between bilingual and monolingual groups. Also, novel information was recognized similarly to familiar information and none of the groups was different in this regard.
Keywords: Novelty effect, memory, bilingualism, language integration.
1 Psychology C, Spring 2006. Supervisor: Reza Kormi-Nouri.
People use language to classify and organize information they get from their surrounding environment. One proof of this is that, most memory strategies, are related to language. Memory and language act as a network that relates concepts to other concepts and stores
information about them. A concept, in turn, is related to all the words that can be used to describe it, the meaning of those words and how to pronounce or spell them. This is known as language
integration (Francis, 1999). Language integration is even more complex when words from more than one language are involved. This is the case for bilinguals.
What does it mean to be a bilingual? The most liberal definition of bilingualism states that bilinguals are exposed to two or more languages. According to the most conservative definition, a bilingual is equally proficient in two or more languages. This study will adhere to a moderate definition of bilingualism. According to this definition, bilingualism is the use of two or more languages in everyday life and bilinguals are those individuals who need two or more languages in everyday life (Grosjean, 1992). This implies regular use and competence in two or more languages in reading, listening, writing and speaking.
Previous literature reveals that traditionally, bilingualism has been seen as a cause for delay in language learning and lower performance in verbal tasks. Empirical data points out that in language processing tasks, bilinguals receive information about words in two languages, even if information is needed in one language only (Gollan & Montoya, 2002). This inability to turn off the language that is not in use at the time, produces delays in retrieval tasks. This is because the
bilingual subject will have to check that the information received belongs to the language used in the task (Van Heuven, Dijkstra & Grainger, 1998) Other research suggests that the delay is caused
by the languages interfering with each other (Rosselli et al. 2000). Some studies also suggest that bilinguals have a weaker semantic-phonological connection. Since this connection is built by using words and bilinguals use languages in different contexts, they use words in either language less often than monolinguals do (Burke, MacKay, Worthley & Wade, 1991).
On the other hand, more recent research accounts for the positive effects of bilingualism. Higher memory performance of bilinguals has been specially related to long term memory, and is remarkable in tasks that involve automatic processing of information, rather than deliberate (Kormi-Nouri, Moniri & Nilsson, 2003). Other studies show that control over language integration is higher in bilinguals (Bialystok, 1988). Cognitive performance and memory capabilities are also shown to be higher in bilinguals in comparison to monolinguals (Francis, 1999). Furthermore, bilinguals who learn a second language early in life achieve a higher degree of performance in both languages, whereas learning a second language late in life is related to failure in acquiring high skills in that language (Francis,1999). For this reason, there are more “true” bilinguals among children than among adults. Children who learn a second language at an earlier age develop their general verbal capabilities and become proficient in both languages more easily than those who learn the second language later in life.
Episodic And Semantic Memory
Tulving (1983) distinguished between two forms of long-tem memory: episodic memory and semantic memory. Episodic memory is about episodes or events from the personally experienced past. The information in episodic memory additionally represents the relation of experiences events to the rememberer´s personal identity as it exists in subjective time and space ( for example when and where you had your graduation exam). Semantic memory concerns general knowledge of the World. People share in semantic memory and is not related to time and place (for example, knowledge about capital cities of the World).
The “Novelty Encoding Hypothesis”
The “Novelty Encoding Hypothesis” (Tulving & Kroll, 1995) suggests that performance in encoding information in long term memory is higher for novel than for familiar items. The study supporting this hypothesis had two phases where subjects had to make decisions whether a set of nouns belonged to the category “living”, meaning living beings, or “non-living” meaning non living beings. Half of the nouns were presented in Phase 1 and half of them were new. Both phases were identical except for that in Phase 2, subjects were also told to memorize the nouns. The results showed that subjects recognized more novel nouns, presented at the time of Phase 2 than familiar ones, presented in Phase 1. This was called the novelty effect (Tulving & Kroll, 1995). Further studies about the novelty effect found that specific subcortical and cortical regions of the brain respond more actively to novel than to familiar information (Tulving, 1994, Habib & Lepage 2000).
Despite this evidence, there was criticism of the novelty effect. Some critics suggested that the novelty effect in Tulving´s study was caused by source discrimination and argued that
participants did not know if the information came from Phase 1 or Phase 2 (Dobbins, Kroll, Yonelinas & Liu,1998). Other critics of the Novelty Encoding Hypothesis pointed out that the results in Tulving´s study could be explained by rehearsal strategy account (Maddox & Estes, 1997) or restricted to a particular set of subjects or conditions, thus questioning the generalizability of the novelty effect (Greene, 1999).
To provide support for Tulving´s Novelty Encoding Hypothesis, a study was built up using different materials such as verbs and nouns, different encoding conditions such as a memory task in Phase 1, and a frequency judgement task in Phase 2 (Kormi-Nouri, Nilsson & Ohta, 2005). Thus, this new study made an attempt to avoid the methodological problems of Tulving´s study. Results provided even stronger support for the Novelty Encoding Hypothesis.
cognitive performance in bilinguals on one side and about the novelty effect on the other, but there is no knowledge about how the novelty effect occurs in bilinguals. Specifically, there is no research about the novelty effect in children since previous studies used adult subjects.
The purpose of this study was first to replicate the study of Kormi-Nouri et al. (2005) and second explore whether novelty effect occurs in children as it does in adults. The study also aims to find out whether novelty effect differs between bilingual and monolingual children. In addition, the present study examines if gender influences the novelty effect. The study was carried out on
monolingual and bilingual boys and girls in Swedish schools. The experiment was divided into two phases. Phase 1 was used to measure the performance of the groups in remembering and
recognizing a list of words. In Phase 2, individuals had to make a decision tasks with words and recognize a list of new words and words from the first phase.
Method Participants
Participants were 11 to 14-year-old pupils of both genders in the city of Örebro. The schools were located in peripheral neighbourhoods of the city and were characterized by having high rates of bilingual pupils. Bilingual and monolingual students were selected from the same schools. This selection was a way to ensure that there were more socioeconomic similarities between Swedish-born pupils and those with an immigrant background.
Since pupils were underage, it was required to obtain their parents permission for them to participate in the study. A total of 200 pupils were handed out permission formulars (see appendix 1) and information sheets about the study and asked to return these signed by their parents if they wanted to participate in the study. Participants received no compensation for their participation.
A 26% of the pupils (N = 53) returned the permission formular and volunteered to
participate in the study. Two pupils returned the formular signed as “Do not want to participate in the study” and 73% (N = 146) did not return the formular. Two bilingual participants were not present at the school at the time of the study, and therefore did not take part in it. Thus, the total number of participants in the experiment was 49.
57.1% (N = 28) of the participants reported to be bilingual, meaning that they spoke
Swedish and another language with members of their family (Age mean = 12.36, SD = 0.83 , 61.9% boys, 38.1% girls) and 42.9 % (N = 21) reported to be monolingual, meaning that they spoke only Swedish with the members of their family (Age mean = 12.71, SD = 1.10, 42.9% boys, 57.1% girls)
The bilingual group was composed by pupils of different cultural background (32.1 % Assyrian (N = 9), 17.9 % Arabic (N = 5), 14.3 % Kurdish (N = 4), 10.7 % Somali (N = 3), 3.6 % Turkish (N = 1) , 3.6 % Romanian (N = 1), 3.6 % Polish (N = 1), 3.6 % Russian (N = 1), 3.6 % Tigrinya (N = 1). The monolingual group consisted of Swedish-only speakers. In order to make sure that bilinguals and monolinguals were equivalent in verbal fluency and could be matched,
participants were asked to do a fluency test that consisted in writing as many words starting with B or S as possible within five minutes. After the fluency test, participants had to rate themselves in talking, reading, listening and understanding Swedish. These were practical methods of evaluating the degree of verbal fluency in each group (Moniri, 2006). Furthermore, bilingual participants had to answer a self rating questionnaire in talking, reading, listening and understanding, and a
questionnaire of five questions about their mother tongue. This was used to measure the degree of bilinguality and ensure that the bilingual group adjusted to the definition of bilingualism used in this study.
Materials and procedure
The participants were tested in groups of four to five pupils each time. To control for external factors, the participants were divided into two equal groups (there was aproximately the same
number of monolinguals and bilinguals in each group), and counterbalancing was applied. There were two versions of each test. Half of the participants received one version, and the other half received another version. Total time of the test was between 25 and 30 minutes and was performed during school time. The participants moved to a room where they could not be disturbed. They were asked to bring with them a blank sheet and a pen. Instead of using their names, participants were given a personal code and asked to write it in each sheet that they received. The participants were also asked to answer as honestly as possible during the test.
The material used for the test was a cassette tape with the recorded wordlists for the first phase of the test (see appendix 2), a cassette player, series of sheets containing both instructions and wordlists and blank sheets brought by the participants.
The test was divided into two phases, and there was an interpolated task between each phase. Each phase consisted of three stages
Phase 1
Stage 1. The researcher played a tape with the following instructions: ”You are going to listen to a list of words. Try to repeat them mentally. Later on, you will be given a test and you will have to recognize these words”. After this, the list of words was presented (see appendix 2). The list contained 20 words that came from a list of common words in Swedish language (Allen, 1972 ). The words were nouns characterizing material, concrete objects. Word length was controlled by limiting syllables in the words to between one and three. The rate of presentation was at the most two seconds for each word, followed by a two second silence between words. During this stage, participants were not allowed to write down anything.
Stage 2. The participants were presented a sheet with the instruction: ”Write as many capital cities of the world as you can” (see appendix 3). The time for doing the task was three minutes. Copying items from another participant was not allowed. At the end of the stage, participants were asked to check whether they had written their personal code on top of the sheet and then the researcher collected the sheets..
Stage 3. Participants were given a test list containing both the 20 words presented in Stage 1 and 20 new ones used as distractors (see appendix 4). The words were mixed and presented in random order. Participants were asked to identify which words were presented in the study list from Stage 1 and to mark them with “yes” and which words were not and to mark them with “no”. At the end of the stage, the researcher collected the sheets.
Interpolated task
Between Phase 1 and 2, participants were presented a sheet with the instruction: ”Write as many animals as you can” (see appendix 5). The participants did the task independently from each other, that is, copying was not permitted. They had five minutes to complete the task. At the end of the stage, the researcher collected the sheets. The purpose of using this interpolated task was to have an interval of time between Phase 1 and Phase 2. Furthermore, this test was used as an indicator of semantic memory.
Phase 2
Stage 1. The participants were given a sheet with a set of instructions and a list of 20 words (see appendixes 6 and 7). Half of the words were selected from the study list in Phase 1, the other half were new. In this stage, participants had to rate each word using a scale between 1 to 5, according to how familiar that word was to the person in everyday life. 1 meant that the word was never used in their everyday life, while 5 meant that the word was used very often in every day life. The researcher did not mention at any moment that those words had to be memorized or recognized, but incidental (non-voluntary) learning was expected. At the end of the stage, the researcher
collected the sheets.
Stage 2. The participants were presented a sheet with the instruction: ”Write as many words starting by B and S as you can” (see appendix 8) . The reason for choosing the letters B and S is because words beginning with these letters have intermediate frequencies in the Swedish language (Kormi-Nouri, Moniri & Nilsson, 2003 ) The time for doing the task was three minutes. Copying items from other participants was not allowed, and the participants did the task individually. At the end of the stage, the researcher collected the sheets.
Stage 3. Participants were given a test list containing both the 20 words presented in stage 1 and 20 new ones used as distractors (see appendixes 9 and 10). This 20 new words were completely new, and had not been presented at any time during the study. As in Phase1, all words were
presented in random order and participants were asked to identify which words were presented in the study list from Stage 1 (by marking them with “yes”) and which words not (by marking them with “no”). Besides this, participants also had to indicate if they were sure, unsure or guessing their answers. At the end of the stage, the researcher collected the sheets.
Post-Test questionnaire
After the test, Participants were given a questionnaire where they had to rate themselves in their knowledge of Swedish, and for biliguals also in knowledge of their mother tongue, in writing, listening/ understanding, reading and speaking. A scale from 1 to 5 was used for each item , where 1 was equal to very bad and 5 was equal to very good. After this, there were six (yes/no) questions for the bilingual participants (see appendix 11).
These questions were used to examine to what extent bilinguals used their mother tongue in everyday life. The purpose was to measure the proficiency in the mother tongue in comparison to proficiency in Swedish in the bilingual group. This would represent the degree of bilingualism in the sample used for the study (Moniri, 2006).
Results
To ensure that monolingual and the bilingual groups were comparable, and results were not caused by differences in general memory and verbal capabilities, the groups were tested in two ways: level of competence in the Swedish language and word generation in Swedish. Since the tests,
instructions and items were in Swedish, it was examined whether both groups had a similar level of proficiency in Swedish language and were equally verbal fluent. For this, two measures were used: the self rating test in use of Swedish (in writing, listening/understanding, reading and speaking) in a scale from 1 to 5 and the word generation task using the letters B and S. Additionally, bilinguals had to rate themselves in their use of their mother tongue using the same method.
An ANOVA was used for the analysis of the data. The one way ANOVA analysis showed that there were no differences between monolinguals and bilinguals with respect to the self rating of their capabilities in the Swedish language. On the other hand, bilinguals rated themselves as less proficient in their mother tongue than in Swedish, df(1, 47), Mse = 2.96, F= 1,18 and P > 0.20. See Table 1.
Table 1. Self rating of the participants in both Swedish and mother tongue. Maximum score is 5
Group Writing Listening/understanding understanding Reading Speaking Monolingual (Swedish) 4.58 (SD=0.60) 4.57 (SD=0.51) 4.71 (SD=0.46) 4.86 (SD=0.36) Bilingual (Swedish) 4.53 (SD=0.88) 4.36 (SD=0.91) 4.53 (SD=0.88) 4.50 (SD=0.88) Bilingual (Mother tongue) 2.75 (SD=1.40) 3.79 (SD=1.23) 2.43 (SD=1.32) 3.79 (SD=1.23)
To explore further how proficient bilinguals were in their mother tongue,a post test
questionnaire was used with five Yes/No questions about the use of the mother tongue in everyday life. For each question that participants answered with “yes”, they were given one point. Maximum score was five points and would indicate that a true bilingual that is highly proficient in their mother tongue. Minimum score was 0 (no question answered with “yes”) which would mean that a
bilingual has very little knowledge of his/her mother tongue. Average score was 2.5.
An analysis of the items of the post-test questionnaire showed that the mean score in this questionnaire was under average (m = 1.86, SD = 1.35). This implies that the bilinguals in this study had a low level of proficiency in their mother tongue.
As mentioned before, word generation is also a good indicator of the verbal fluency of the individual. An ANOVA analysis of the number of words generated by the participants suggested that there were no differences in the number of words generated by bilinguals and the number of words generated by monolinguals, df(1, 47), Mse = 53.57, F= 0.02 and P > 0.90. See Table 2.
Table 2. Number of words starting by the letters B and S generated by bilinguals and monolinguals
B S Total Bilingual 10.86 SD=5.68 12.96 SD=6.40 11.91 SD=6.04 Monolingual 11.24 SD=5.28 12.19 SD=4.73 11.71 SD=5.0 TOTAL 11.05 SD=5.48 12.58 SD=5.56
The ANOVA analysis suggested that there was no interaction effect between groups
(monolingual/bilingual) and letter (B/S): df(1,47) , Mse = 14.63, F = 0.55 and P > 0.40. However, it should be noted that there was a tendency effect for both bilingual and monolingual to write more words starting with S than B, df (1, 47), Mse = 14.63, F= 3.84 and P < 0.06.
Measure of Memory Performance
The next step was to explore the memory performance in bilingual and monolingual groups. In line with the study of Kormi-Nouri et al.(2003), both episodic and semantic memory were tested in this study. To measure performance of episodic memory, I used the memory tasks in Phase 1 and 2 where participants had to recognize a list of 40 words (20 words that had already been presented before and 20 words were new word). The total number of items were recognized correctly (hits) and items were recognized wrongly as correct items (false alarms) for each group were calculated.
In hit rates data, the results showed that there was no difference between groups in Phase 1, df (1, 47) , MSe = 12.48 , F= 0.12 and P > 0.70. Also, there was no difference between groups in Phase 2, df (1,47) Mse 13,72, F=4.01 and P > 0.05.
In hits – false alarm data, the results showed that there was no difference between groups in Phase 1, df (1, 47) , MSe = 16.73 , F = 0.16 and P > 0.60. Also, there was no difference between groups in Phase 2, df (1,47) Mse = 38.82, F= 0.80 and P > 0.30. See Table 3.
Table 3. Hits and false alarms for monolinguals and bilinguals in Phase 1 and Phase 2
PHASE 1 Hits Hits – False alarms
Monolingual 70 % 47 %
Bilingual 72 % 44 %
PHASE 2 Hits Hits – False alarms
Monolingual 83 % 60 %
For semantic memory, I examined the tasks of writing as many animals and capital cities of the world as possible during a limited time (3 minutes). The results showed that there was no difference in the task of capital cities, df(1, 47), Mse = 31.09, F= 0.05 and P > 0.80. Also there was no difference between the two groups in the task of writing animal names, df (1,47), Mse = 128.87, F =1.35 and P > 0.20.
Table 4. Capital cities and animal names produced by monolinguals and bilinguals
Capital cities Animals
Monolinguals 11.48 SD = 4.64 33.52 SD = 9.84 Bilinguals 11.82 SD = 5.83 29.71 SD = 11.10
From results in both semantic and episodic memory, it can be concluded that there was no difference between bilingual and monolingual children in memory performance.
Measure of the Novelty Effect
To search for the novelty effect, it was necessary to compare the number of novel and familiar
words recognized in Phase 2 of the experiment. If novelty effect really occurred, then the number of novel words that were recognized should be higher than for familiar words.
To examine this, an analysis of multifactorial ANOVA with two factors was used: “Group” (monolingual/bilingual), and “Item” (novel/familiar). Analysis of the single factor “item” showed that although novel items (78%) were recognized better than familiar items (73%), this difference did not reach to significance level, df (1, 47), Mse = 5.95, F = 1.11 and P > 0.30. See Table 5. Table 5. Percentage of novel and familiar items recognized by participants.
Novel Familiar Total
Bilinguals 72 % 72 % 72 %
Monolinguals 85 % 74 % 79 %
TOTAL 78 % 73 %
Analysis of the factor “group” revealed that there was no significant difference between monolinguals and bilinguals df ( 1,47), Mse = 7.71,, F = 1.70 and P > 0.20. An analysis of
interactions between the two factors (group and item) did not either show any significant difference between monolinguals and bilinguals df ( 1,47), Mse = 5.95,, F = 1.11 and P > 0.30.
Simple effect comparisons with ANOVA showed that there was group difference in novel items compared to familiar items. That is, the monolingual group had better word recall for novel items than the bilingual group, df (1,47), Mse = 19.11, F = 4,08 and P < 0.05, whereas, for familiar items, there was no such difference, df ( 1,45), Mse = 11.23, F = 1.69 and P > 0.20. It should be mentioned that gender was also included as a factor in the ANOVA analysis, but the results did not show any difference for boys or girls neither in novel items df (1,47), Mse = 1.23, F = 0.24 and P > 0.62 or familiar items df (1,45), Mse = 0.03, F = 0.00 and P > 0.98
Discussion
This research was modelled after a study by Kormi-Nouri et al. (2003). That study argued that bilingualism has positive effects on cognitive performance, such as episodic and semantic memory. The purpose of this study was to replicate the study of Kormi-Nouri et al. (2005) and to examine how novelty effect occurs in bilingual children and at the same time provide support for
the positive effects of bilingualism
In line with previous literature, in this study it was expected that since bilingual children are more creative and better in problem solving in new situations (Francis,1999), they would also be better in recognition of novel items. However, I did not find any positive effects of bilingualism. Monolinguals and bilinguals showed equal performance in memory tasks, both in episodic and semantic memory.
The results of this study did not show any positive effect in bilinguals compared to
monolinguals.One explanation for the inconsistency between this research and that of Kormi-Nouri (2003) is that whereas in (Kormi-Nouri, Moniri & Nilsson, 2003) study used verbal and action sentences, in this study I only used words. It is possible that bilinguality effect would be more observable for related and integrated items such as sentences and not for single and unrelated words.
In Kormi-Nouri et al (2003) study, it was found that bilinguality effect was even more observable for action sentences compared to verbal sentences. It is explained that there is more episodic integration in action sentences compared to verbal sentences.
Another explanation for this inconsistency is that results would be related to the diversity within the subjects of the bilingual group. Whereas Kormi-Nouri et al (2003) used one
homogeneous group of Persian bilingual subjects, this study used different subgroups (Arab, Bosnian, Somali, Kurdish, Turkish, Assyrian, Russian, Polish, Tigrinya and Romanian) of subjects in the bilingual group. All this can be seen as a hinder for the bilinguality effect, since it causes a high variability in the data obtained.
Furthermore, bilingual participants in this study were much more proficient in Swedish than in their mother tongue, and according to the definition of Grosjean (1992), it is possible that
bilinguals in this study were not “true” bilinguals. This would be the reason why the bilinguality effect was not observed in the present study.
In my knowledge, there was no existing research about the novelty effect in children compared to adults, and in bilingual children in comparison with monolingual children. Therefore the second goal of this study was to find out if children, like adults, recognize new information better than familiar information, and if there is any difference between bilingual and monolingual children in this sense. The results showed that children do not recognize new information better than familiar, contrary to what has been shown for adults. Moreover, monolingual children seem to recognize better novel information than bilingual, although this difference was not significant in this sample.
A speculative explanation is that the absence of novelty effect in children in comparison to adults can indicate that perhaps the novelty effect appears as a consequence of language and memory development. General development of memory capabilities and strategies occurs between 6 and 12 years (Schneider, 2002) and after puberty, cerebral hemispheres lose plasticity and become more specialized (Lenneberg, 1967). If we take into account the age of the participants (mean = 12,5 years old), where hemisphere specialization has not yet appeared, it seems reasonable to think that novelty effect is affected by hemisphere specialization.
Nevertheless, there is still need for an explanation for the fact that monolinguals seem to recall novel items better than the bilingual group, even if the difference found in the study was not significant. Again, a speculative explanation of this could be that hemisphere specialization occurs later in bilinguals than in monolinguals, and therefore novelty effect appears sooner in
monolinguals than bilinguals. However, further studies are needed to investigate these speculations. Some issues in this study can be improved in future research. One of them is the size of the sample and the number of participants can be increased to make results more reliable. 200 pupils were asked to participate in the study, but only 50 volunteered to participate. External validity of this study is limited, since the number of participants does not allow to generalize the results of the study to the general population, as it is too low.
The high experimental dropout rate can be regarded as a threat to the internal validity of the study, but there are reasons to think that this does not cause any bias in the results. It is believed that the reason for the high number of students that did not turn in the permission formular is that they simply did not remember about it, or found it embarrasing to give the paper to their parents and bring them back to school: a number of pupils showed interest to participate in the study, but they had not brought back the permission formular, and therefore were not allowed to participate.
. The fact that experimental dropout rate was similar in all schools supports this idea, and suggests that the results of the study were not affected by an hypothetical bias in the sample.
One way to increase the number of participants in future studies would be to make it easier for the participants to take part in the study, for example, using the permission of the school principal instead of the permission of parents to allow pupils to participate in the study.
Another way of increasing the number of participants would be to better control the delivery of the permission formulars to the parents or arranging a meeting and informing them about the study. An additional way of increasing the number of participants would be to to introduce a reward for the participants.
An issue that can threaten the internal validity of the study is that there was a high
variability in the data obtained from the study. Future studies can overcome this issue by limiting the bilingual group to subjects that have all the same mother tongue, and are all of the same age.
However, this study is original in the sense that it does an important contribution to the knowledge about novelty effect, because it uses a sample of children, contrary to previous studies about the novelty effect, that use a sample of adults. Examining memory and language mechanisms might have implications in education, linguistics, language learning and cognitive psychology. Knowledge about memory and language can be specially useful in environments where
bilingualism is present.
Another positive factor of this study is the use of instruments from previous studies of bilingualism and memory and the use of counterbalancing for both bilingual and monolingual groups. This can ensure the validity and internal consistency of the measures used in this study, and helps controlling external factors.
This study can also encourage further research in how novelty effect develops as a consequence of language learning and hemisphere specialization, and how it is affected by bilingualism.
References
Allen, S. (1972). Tiotusen i topp. Stockholm: Almqvist & Wiksell
Bialystok, E. (1988). Levels of bilingualism and levels of semantic awareness. Developmental psychology 24, 560-567.
Burke, D., MacKay, D., Worthley, J. & Wade, E. (1991).On the tip of the tongue: What causes word finding failures in young and older adults? Journal of Memory and Language, 30, 542-579.
Dobbins, I., Kroll, N., Yonelinas, A., & Liu, Q. (1998). Distinctiveness in recognition and free recall: The role of recollection in the rejection of the familiar. Journal of Memory and Language, 38, 381-400.
Francis, W. S. (1999). Cognitive integration on language and memory in bilinguals: Semantic representation. Psychological Bulletin 125, 193-222.
Gollan, T. H. & Montoya R. I. (2002). Semantic and letter fluency in Spanish-English Bilinguals. Neuropsychology 16, 562-576.
Grosjean, F. (1992). Another view of bilingualism. In R. Harris (Ed), Cognitive processing in bilinguals (pp. 51-62). Amsterdam: Elsevier.
Greene, R.(1999). The role of familiarity in recognition. Psychonomic Bulletin & Review, 6, 309-312.
Habib, R & Lepage, M. (2000). Novelty assesment in the brain. In E. Tulving (Ed.), Memory, consciousness, and the brain: The Tallin conference (pp. 265-277), Philadelphia, PA: Psychology Press
Kormi-Nouri, R., Moniri, S., & Nilsson, L.-G. (2003). Episodic and semantic memory in bilingual and monolingual children. Scandinavian Journal of Psychology, 44, 47-54.
Kormi-Nouri, R., Nilsson, L.-G., & Ohta, N. (2005). The Novelty effect: Support for the Novelty-Encoding Hypothesis. Scandinavian Journal of Psychology, 46, 133-143.
Lenneberg, E. (1967). Biological foundations of language. NY : Wiley and sons.
Maddox, W. T. & Estes, W. K. (1997). Direct and indirect stimulus frequency effects in
recognition. Journal of Experimental Psychology: Learning, Memory & Cognition, 23, 539-559.
Moniri, Sadegheh (2006). Bilingual memory: A lifespan approach. Stockholm university
Rosselli, M., Ardila, A., Araujo, K., Weekes, V. A., Caracciolo, V., Padilla, M., & Ostroaky-Solis, F. (2000). Verbal fluency and repetition skills in healthy older Spanish-English bilinguals. Applied Neuropsychology, 7, 14-24
Schneider, W. (2002). Memory development in childhood. In U. Goswami (Ed.), Blackwell
handbook of cognitive development. Blackwell handbooks of developmental psychology (pp. 236-256). Malden, MA: Blackwell publishing.
Tulving, E. (1983). Elements of episodic memory. New york: Oxford University Press. Tulving, E., Kapur, S., Craick, F. I. M., Moscovitch, M., & Houle, S. (1994). Hemispheric
encoding/retrieval assymetry in episodic memory: Positron emission tomography findings. Proceedings of the National Academy of Sciences, 91, 2016-2020
Tulving, E. & Kroll, N. (1995). Novelty assessment in the brain and long-term memory encoding. Psychonomic Bulletin & Review, 2, 387-390
Van Heuven, W.J.B., Dijkstra, T. & Grainger, J. (1998). Orthographic neighbourhood effects in bilingual word recognition. Journal of Memory and Language, 39, 458-483.
Grup1 och 2
Ange för varje ord om det förekom i listan som tidigare presenterades. Ringa in ditt svar (JA) eller (NEJ) samt om du är (S)äker, (O)säker eller (G)issar.
(S)äker (O)säker (G)issar
Stol JA NEJ S O G Mössa JA NEJ S O G Cykel JA NEJ S O G Ståltråd JA NEJ S O G Hjul JA NEJ S O G Burk JA NEJ S O G Duk JA NEJ S O G Äggklocka JA NEJ S O G Kniv JA NEJ S O G Glas JA NEJ S O G Byxor JA NEJ S O G Brev JA NEJ S O G Paket JA NEJ S O G Flagga JA NEJ S O G Plånbok JA NEJ S O G Album JA NEJ S O G Bil JA NEJ S O G Salt JA NEJ S O G Piano JA NEJ S O G Bus JA NEJ S O G Handske JA NEJ S O G Tidning JA NEJ S O G Ljus JA NEJ S O G Glödlampa JA NEJ S O G Skåp JA NEJ S O G Vagn JA NEJ S O G Bord JA NEJ S O G Tvål JA NEJ S O G Lapp JA NEJ S O G Bild JA NEJ S O G Paraply JA NEJ S O G Ram JA NEJ S O G Citron JA NEJ S O G Legobit JA NEJ S O G Kula JA NEJ S O G Bok JA NEJ S O G Hål JA NEJ S O G Kort JA NEJ S O G Tåg JA NEJ S O G Säck JA NEJ S O G
WORDS PRESENTED IN PHASE 1 (PLAYED ON A TAPE)
Group 1 : Lapp, Byxor, Piano, Citron, Burk, Bild, Bok, Flagga, Säck, Buss, Legobit, Paket, Glas, Album, Brev, Tåg, Handske, Äggklocka, Glodlampa, Skåp.
Group 2: Tvål, Ram, Cykel, Kort, Stol, Kula, Paraply, Vag, Hål, Bord, Salt, Duk, Bil, Ståltråd, Plånbok, Hjul, Tidning, Mössa, Ljus, Kniv.
Skriv hur många huvudstäder i Världen du kan
Skriv hur många ord som börjar med B eller S du kan.
KOD:
A) Din uppgift här är att på en fem-gradig skala, där 1 = mycket dålig och 5 = mycket bra, skatta hur bra du är på SVENSKA SPRÅKET (ringa in).
Att Skriva 1 2 3 4 5
Att lyssna/förstå 1 2 3 4 5
Att läsa 1 2 3 4 5
Att tala 1 2 3 4 5
B) Om du är tvåspråkig, skriv vilket språk som är ditt hemspråk ...
Din uppgift här är att på en fem-gradig skala, där 1 = mycket dålig och 5 = mycket bra, skatta hur bra du är på HEMSPRÅKET om du är tvåspråkig (ringa in).
Att Skriva 1 2 3 4 5
Att lyssna/förstå 1 2 3 4 5
Att läsa 1 2 3 4 5
Att tala 1 2 3 4 5
Svara med Ja/Nej på följande frågor (ringa in):
Jag tittar på Tv/lyssnar på radio på mitt hemspråk (JA) (NEJ)
Jag ser mig själv lika bra på Svenska som på mitt hemspråk (JA) (NEJ) Ibland så skäms jag över mitt hemspråk (JA) (NEJ)
Jag skulle kunna skriva ett prov på mitt hemspråk och få ett bra betyg (JA) (NEJ)
