In the beginning of 20th century research on private speech was actively pursued by J. Piaget, L.S. Vygotskiy and their collaborators (see, for example, the works of R.E. Levina). An enormous scope of empirical and theoretical material was collected at that time. The works of P.Ya. Galperin demonstrate that social speech and inner speech are important instruments of transition from external action to internal.
A work by L.F. Obukhova and I.A. Korepanova (2005) demonstrates that speech is represented differently in different levels of zone of proximal development. However it was not a subject for specific consideration. The following research brings into focus of attention the interconnection between speech (uppermost private speech) and levels of zone of proximal development.
There is practically no other research that compares the features of private speech in different language cultures. It appears urgent to investigate how the language culture itself supports or diminishes effectiveness of private speech (especially in a situation of cooperation).
The aim of the study is to explore the cross-cultural differences and the ontogenetic changes in the structure and functions of private speech in a problem solving context.
A set of hypotheses were generated pertaining to the two domains of enquiry-to the within-group ontogenetic changes of the occurrence, structure and semantic content of private speech and to the between groups cross-cultural differences in these facets of private speech.
It was hypothesised that within both the Russian and the English-speaking group.
- Private speech will demonstrate a curvilinear trend incorporating an increase followed by a subsequent decline.
- Inner speech will exhibit linear increase.
- Social speech will manifest linear decrease.
- There will be significant age-related changes in the structure of private speech.
- There will be significant age-related changes in the content of private speech.
- The shift of competencies from ZPD to ZAD will be associated with a heightened use of private speech.
- Task-orientated contents of private and of social speech together with inner speech will be significant predictors of current task performance.
- There will be significant cross-cultural differences in the structure of private speech.
- There will be significant cross-cultural differences in the content of private speech.
The participants will include Russian and English speaking children from 5 years to 10 years of age who will be recruited from pre-primary and primary schools in Moscow, Russia and in Johannesburg, South Africa.
Participants will be divided into 2 cultural/language groups (Russian and English speaking) and in 4 age groups as follows:
Age group 1- preschool children (age around 5 years)
Age group 2- pre-primary school children (age around 6 years)
Age group 3 - grade one children (age around 7 years)
Raven’s progressive matrices (Form A and B) will be used as a problem-solving task.
Note: here I am referring to different versions, not to the different series within the test.
Task completion will include two sessions. In session 1 participants will be invited to complete Raven’s progressive matrices (Form A) under “alone/exam/class test” condition and no talk will be “permitted” . Under this condition it is presumed that the test score will be a measure of the child’s zone of actual development (ZAD). After a short break (straight after), during session 2, they will be asked to complete Form B. Children from each age group will be randomly divided into triads. The experimenter will be present during the two sessions but only in the second session he/she will be available to assist the participants if they requested help, asked questions and/or required any additional clarification on the task. Under this condition (work in triads) the test score will reflect the position of the child’s capacities within his/or her zone of proximal development (ZPD). The gain in the scores (difference between the test score in session 1 and session2) will reflect how far the competences had shifted forward when self-talk (private speech), and talk to others-either peers or instructor (social speech) have been used as a problem-solving tool.
The participants will be instructed that the completion of the task during session 1 and session 2 is not for the purpose of any school/academic related matters, that the results are confidential, and that they should feel free to talk openly during the completion of the task (in session two only), if they wished to do so.
The task completion sessions (1and 2) will be video recorded. Speech utterances produced by the children will be transcribed and then coded using a comprehensive coding system.
Each utterance will be coded in terms of the type of speech it belongs to, its structure and its semantic content.
The variable speech type will include categories private, social and inner speech which will be operationalized as follows:
- Social speech category will include utterances that were addressed to another (either peer or adult);
- Private speech will include any other overt utterances;
- Inner speech will comprise inaudible muttering and lips and tongue movements.
The variable utterance structure will include the categories:
- Complete utterance structure: Utterances will be considered to have been complete if they were grammatically correct and/or comprehensive to non-present other.
- Fragmented or abbreviated utterance structure which will be inclusive of
- Predicate-mode utterance structure- if the utterance contained predicate part only
- Subject-mode utterance structure- if the utterance contained the subject part only
The functions of private speech may not be measured directly; they will be extrapolated from the following indicators:
- role in the transition of competencies from ZPD to ZAD;
- significance for predicting current task performance.
The variable utterance content will include the categories:
- Describing task activity (cognitive and metacognitive function)
- Describing task material (cognitive and metacognitive function)
- Requests, questions, commands (cognitive and metacognitive function)
- Affect expressions and sounds (emotional release and emotional self-regulation)
- Describing peer’s task (cognitive and metacognitive function)
- Task-irrelevant content (indeterminate function)
Mean frequencies measures (utterances per minute) will be calculated for each of the variables. Data will be analysed through the following statistical methods:
1. For Hypotheses 1,2,3 a one way MANOVA will be performed separately for the Russian and the English speaking group where age group group will feature as the independent variable and private, social and inner speech- as the dependent variables;
2. For Hypotheses 4 and 5 a one way MANCOVAs will be performed with age group featuring as the independent variables. In the first run of MANCOVA the dependent variables will be utterance structures of private speech and the covariates will be the corresponding utterance structures of social speech and inner speech. In the second run, the dependent variables will be utterance contents of private speech and the covariates will be the corresponding utterance contents of social speech and inner speech.
3. For Hypotheses 6 and 7 Multiple Regression analyses will be performed separately for the Russian and the English speaking groups. In the first run the outcome variable will be the difference in the task performance scores between Session1 and Session 2 and the predictor variables will be private, social and inner speech. In the second run the outcome variable will be the task performance score in Session 2 and the predictor variables will be the task-orientated contents of private and of social speech together with inner speech.
1. For Hypotheses 1 and 2 a one way MANCOVAs will be performed with language group featuring as the independent variables. In the first run of MANCOVA the dependent variables will be utterance structures of private speech and the covariates will be the corresponding utterance structures of social speech and inner speech. In the second run, the dependent variables will be utterance contents of private speech and the covariates will be the corresponding utterance contents of social speech and inner speech.