SciELO - Scientific Electronic Library Online

 
vol.61 issue4Rett syndrome: clinical and epidemiological aspects in a Brazilian institutionMemory complaints and test performance in healthy elderly persons author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand

Journal

Article

Indicators

Related links

Share


Arquivos de Neuro-Psiquiatria

Print version ISSN 0004-282XOn-line version ISSN 1678-4227

Arq. Neuro-Psiquiatr. vol.61 no.4 São Paulo Dec. 2003

http://dx.doi.org/10.1590/S0004-282X2003000600005 

Brazilian standardised norms for a set of pictures are comparable with those obtained internationally

 

A padronização brasileira para um conjunto de figuras é comparável à obtida internacionalmente

 

 

Sabine PompéiaI; Mônica Carolina MirandaI; Orlando Francisco Amodeo BuenoII

Departamento de Psicobiologia da Universidade Federal de São Paulo (UNIFESP), São Paulo SP, Brasil
IGraduate Student
IIProfessor

 

 


ABSTRACT

Snodgrass & Vanderwart (1980) standardized a set of 260 pictures in the USA for use in studies of cognitive processes that employ pictured objects as laboratory analogues of object themselves. Since then similar norms for this set were obtained in Britain, Spain, Japan and Iceland and a larger set of 400 pictures (including the original 260: Cycowicz et al., 1997) was studied in France and Brazil. The present article provides a comparison of the norms obtained in Brazil and internationally. The pattern of correlations among the Brazilian and other standardizations were equivalent to that previously observed: despite pictures being judged to be of similar familiarity and visual complexity (high positive correlations), name agreement was less correlated, possibly due to differences in the languages spoken in each country and/or in the sample size used in each study. Results confirm the adequacy of the Brazilian norms.

Key words: picture, naming, familiarity, visual complexity, norms, standardization.


RESUMO

Snodgrass & Vanderwart (1980) padronizaram um conjunto de 260 figuras nos EUA para uso em pesquisas de processos cognitivos nas quais figuras de objetos são utilizadas como análogos laboratoriais de objetos propriamente ditos. Desde então normas similares foram obtidas no Reino Unido, Espanha, Japão e Islândia, e um conjunto de 400 figuras (incluindo as 260 originais: Cycowicz et al., 1997) foi estudado na França e no Brasil. O presente estudo traz uma comparação das normas obtidas no Brasil e internacionalmente. O padrão de correlações entre as padronizações brasileira e as demais foi equiparável ao previamente descrito: apesar das figuras serem julgadas como tendo familiaridade e complexidade visual similares (correlações positivas elevadas), a consistência de nomeação foi menos correlacionada, possivelmente devido a diferenças nos idiomas de cada pais pesquisado e/ou devido ao tamanho da amostra empregada em cada estudo. Os resultados confirmaram a adequação das normas brasileiras.

Palavras-chave: figura, nomeação, familiaridade, complexidade visual, normas, padronização.


 

 

Investigations into the complex cognitive processes involved in naming objects1,2 require that studies be conducted under carefully controlled conditions. These often include pictured objects as laboratory analogues of object themselves. However, no normative data for such stimuli was available until the end of the 1970s, precluding adequate comparisons between studies because they employed different sets of drawings. A turning point in this line of study was the publication of Snodgrass & Vanderwart's3 standardisation of 260 pictures of common objects drawn in black over a white background. These stimuli were drawn so as to follow pre-determined rules that permit evaluation of consistency between them, such as size4, number of details and orientation3. Among the most important aspects of pictures determined in Snodgrass & Vanderwart's3 study were name agreement, or the rate at which objects depicted in the drawings are referred to with the same name, familiarity with the pictured concepts and visual complexity of the drawings. Name agreement is a robust predictor of naming difficulty and is important for studies of naming latency, picture-name matching, recall, recognition and investigations in which verbal coding is manipulated6. Familiarity is an important predictor of picture naming latencies (the more familiar the concepts, the shorter the naming time), while visual complexity affects variables such as naming latency, tachistoscopic recognition threshold and memorability6. Since the pioneering paper of Snodgrass & Vanderwart3 in the USA, similar work has been conducted on the 260 picture-set for British7, Spanish8, Japanese9 and Icelandic10 university students. A larger set of 400 pictures, which includes the original 260, has been studied more recently for both Frenchmen6 and Brazilians11. The number of subjects and of pictures used in each normative paper can be found in Table 1.

 

 

 

 

Correlations between variables in each of the normative studies were conducted in order to determine how they related to each other (Table 3). Measures of name agreement were highly negatively correlated, ratings of familiarity and complexity were modestly correlated, and the remaining comparisons between measures yielded small or non-significant effects. The largest correlations found occurred in the Brazilian study, probably because it used the larger sample (Table 1) and had every subjects rate naming, familiarity and visual complexity of all 400 pictures, thus lowering variance,, which increases correlations. Nevertheless, the pattern of correlations betweenamong measures reported for different cultures is equivalent, supporting the idea that naming, familiarity and complexity are essentially independent and may be assumed to affect different stages during picture processing5.

 

 

Correlations between norms of different cultures were also conducted in some of the above-mentioned papers (Table 2) in order to determine whether ratings from different countries were comparable. This comparison can be carried out because the procedure employed in all studies was essentially the same: subjects were all young university students that were native speakers of the particular language under investigation; were tested in groups (except12, in which individuals were tested alone); were instructed to name the objects depicted on slides or on the computer screen, as well as to judge their familiarity with the concept and visual complexity of the drawings using 5 point scales (1 represented the least familiar and complex). Results showed large significant positive correlations for familiarity and complexity ratings (Table 2). This probably occurs because such measures relate to judgements about the pictures or objects themselves, which are similar irrespective of the language spoken by the person who rates them8. Smaller correlations occur for name agreement, which is not surprising considering that the word or words used to name each object varies across tongues8. Hence, the similar pattern of correlations among naming, familiarity with concepts and complexity of drawings in each normative study, and between these measures in different cultures, confirm the usefulness of the 2603 and 4005 picture-sets as tools for international cognitive research.

The objective of the present study was to determine the adequacy of the Brazilian normative data through the comparison (correlations) of results on name agreement, familiarity and complexity with those of other cultural groups.

 

METHOD

Comparisons between the Brazilian11, North-American3, French6, British7, Spanish8, Japanese9 and Icelandic10 norms were conducted considering the following measures for each picture:

a. Name agreement: refers to the degree to which subjects agree on the name of the picture. Two measures were used: the percentage of subjects who used the modal name and the H index, calculated in the following manner:

This H index takes into account the number of subjects that gives each one of the different names used for the same picture3,5; k refers to the number of different names given to each picture, and the Pi is the proportion of subjects who gave each name. The greater the naming agreement between subjects, the closer the H is to 0.

b. Familiarity: refers to the familiarity of the concept depicted. Scores ranged from 1 to 5 (1=very unfamiliar, 2=unfamiliar, 3=medium, 4=familiar, 5=very familiar).

c. Visual complexity: refers to the amount of lines and details in the drawing. Scores ranged from 1 to 5 (1= least complex; 5= most complex).

Some of the normative studies also evaluated other measures for the sets of pictures such as age of acquisition (through adult estimates of word learning age, a method that is no longer considered a good predictor of this variable10,12), imageability, image agreement, and naming latency. These variables were not established in the Brazilian norms and will therefore not be discussed here.

Statistical analysis involved pictures as units of measure. Thus, comparisons were conducted considering the pictures for which data were available in both the Brazilian and other norms (see Table 1 for information on the number of pictures studied in each culture). Most measures did not show normal distribution or homocedasticity, so comparisons were conducted through non parametric Spearman rho correlations. The level of significance adopted was 0.01 because of the large number of comparisons conducted.

 

RESULTS

Table 4 lists the correlations between the Brazilian data and that of the other countries for which norms were obtained.

 

 

DISCUSSION

Correlations of the Brazilian normative data with that of other cultures showed that despite pictures being judged to be of similar familiarity and visual complexity (high positive correlations), name agreement was less correlated, as previously observed6,8,9.

High correlations between familiarity and complexity ratings probably occurs because such measures relate to judgements about the pictures or objects themselves, which are similar irrespective of the language spoken by the person who rates them8.

In terms of naming, the smallest correlations obtained were between the Brazilians and Japanese, confirming low agreement between norms from Japan and the USA9. Therefore, similarity in the languages spoken seems to influence correlations in naming because they were higher when comparing data from the Brazilian, Portuguese speaking sample, with that of Spaniards8 and Frenchmen6 (both of which use Latin rooted languages), than when comparing Brazilian with Japanese9, British7, and Icelandic10 norms. However, correlations between naming by the British7 and Brazilian samples were surprisingly low, very similar to that of Brazil and Japan, although the language spoken is the same as in North-America3, which had naming ratings that were very similar to those obtained in Brazil. Hence, differences in naming seem to follow some other characteristic that not only the language spoken in the country in which norms are obtained. We suggest that correlations of naming measures are also determined by the sample size, which was small in the British and Icelandic studies (Table 1). Differences in naming may be over or underrepresented if few subjects are used, increasing variability which inevitably lowers correlations. Differently, familiarity and complexity norms did not seem to be in any way related to the sample size used. This is not surprising since ratings of these measures vary only between 1 and 5, while it is impossible to determine how many different names will be attributed to a picture.

The pattern of correlations between the Brazilian and international standardizations thus confirms the adequacy of the Brazilian norms and that data from this country using these pictures can be directly compared to that obtained abroad. Nevertheless, apart from cultural differences, the names attributed to pictures may also differ among subjects who speak the same native language but who inhabit distinct regions or countries, are from different social and educational background, and of different ages. Norms for children were obtained only for the North- Americans5,13 and Brazilians11; information on picture ratings by other age groups are lacking. Also, gender differences in picture attributes have seldom been studied and were not determined in any of the norms cited here, so they may also lead to distinct effects. Hence, pilot studies with the specific population to be investigated should always be conducted so as to determine the adequacy of the norms employed.

 

REFERENCES

1. Johnson CJ, Clark JM, Paivio A. Cognitive components of picture naming. Psychol Bull 1996;120:113-139.        [ Links ]

2. Murtha S, Chertkow H, Beauregard M, Evans A. The neural substrate of picture naming. J Cogn Neurosci 1999;11:399-423.        [ Links ]

3. Snodgrass JG, Vanderwart M. A standardised set of 260 pictures: norms for name agreement, image agreement, familiarity, and visual complexity. J Exp Psychol Hum Learn Mem 1980;6:174-215.        [ Links ]

4. Theios J, Amrhein PC. Theoretical analysis of the cognitive processing of lexical and pictorial stimuli: reading, naming, and visual and conceptual comparisons. Psychol Rev 1989;96:5-24.        [ Links ]

5. Cycowicz YM, Friedman D, Rothstein M, Snodgrass JG. Picture naming by young children: norms for name agreement, familiarity, and visual complexity. J Exp Child Psychol 1997;65:171-237.        [ Links ]

6. Alario F-X, Ferrand L. A set of 400 pictures standardised for French: norms for name agreement, image agreement, familiarity, visual complexity, image variability, and age of acquisition. Behav Res Method Instrum Comput 1999;31:351-552.        [ Links ]

7. Barry C, Morrison CM, Ellis AW. Naming the Snodgrass and Vanderwart pictures: effects of age of acquisition, frequency, and name agreement. Quart J Exp Psychol 1997;50A:560-585.        [ Links ]

8. Sanfeliu MC, Fernandez A. A set of 254 Snodgrass-Vanderwart pictures standardised for Spanish. Behav Res Method Instrum Comput 1996;28:537-555.        [ Links ]

9. Matsukawa J. A study of characteristics of pictorial material. Memoirs of the faculty of Law and Literature: Shimane University 1983. Shimaneken, Japan.         [ Links ]

10. Pind J, Jonsdottir H, Tryggvadottir HB, Jonsson F. Icelandic norms for the Snodgrass and Vanderwart (1980) pictures: name and image agreement, familiarity, and age of acquisition. Scand J Psychol 2000;41:41-48.        [ Links ]

11. Pompéia S, Miranda MC, Bueno OFA. A set of 400 pictures standardised for Portuguese: norms for name agreement, familiarity and visual complexity for children and adults. Arq Neuropsiquiatr 2001;59:330-337.        [ Links ]

12. Morrison CM, Ellis AW. Real age of acquisition effects in word naming and lexical decision. Br J Psychol 2000;91:167-180.        [ Links ]

13. Berman S, Friedman D, Hamberger M, SnodgrassJG. Developmental picture norms: relationship between name agreement, familiarity, and visual complexity for child and adult ratings of two sets of line drawings. Behav Res Method Instrum Comput 1989;21:371-382.        [ Links ]

14. Snodgrass JG, Corwin J. Perceptual identification thresholds for 150 fragmented pictures from the Snodgrass and Vanderwart picture set. Percept Mot Skills 1988;67:3-36.        [ Links ]

 

 

Received 1 April 2003. Accepted 21 June 2003
Study supported by AFIP and FAPESP (grant 1998/15397-1)

 

 

Dr. Orlando F.A. Bueno - Departamento de Psicobiologia-UNIFESP - Rua Botucatu 826/1º andar - 04023-062 São Paulo SP - Brasil. FAX: 55 11 5572-5092. E-mail: ofabueno@psicobio.epm.br

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License