THE LEARNING STRATEGY OF THE TOTAL PHYSICAL RESPONSE: SOME AGE DIFFERENCES JAMES J. ASHER BEN S. PRICE
San Jose State College There is a common belief that children arc better able than adults to leam a foreign language. This belief may be an illusion if children living in a foreign country leam the new language through phy activity while their parents try to leam independently of physical behavior. Using Russian, this study compared the listening comprehension of 8-, 10- and 14-year-old children, and college adults when each S was in physical action during retention tests. The results showed that (a) the adults were superior to the children of any age group af p < .0005; (b) the youngest children, the 8-year-olds, had the poorest retention; and (c) the 10- and 14-year'old children were intermediate between the adults and tJw 8-year-olds. Tbere is an almost irrefutable belief that children are better able than adults to leam a foreign language. This belief may be the result of a common observation that when children live in a foreign country they acquire speaking fiuency of a new language while tbeir parents are retarded in understanding and vocahzing. It may be that children outperform adults in foreign language comprehension because tbe new language is leamcd through play activity in whicb the child makes action responses. For tbe child, the second language tends to be synchronized witb physical responses ("Come on, Sam. Let's ride our bikesl"). The adult, by contrast, tries to manipulate the foreign This study was supported by a research contract from the Personnel and Training Branch of the Office of Naval Research (NONR-4817[00], NR-154257/12-8-64). Reproduction in whole or part is permitted for any purpose of the U.S. government. Appreciation is expressed to Drs. Rose Cinsberg and Cene R. Medinnus for their valuable suggestions in reviewing this paper. Author Asher's address: Psychology Department, Child Research Center, San Jose State CoUege, San Jose, California.
CHILD DEVELOPMENT
language quite independently of physical behavior. The adult tends to be physically static when he receives or transmits the new language ("It's a beautiful day today, isn't it?"). If the child in a foreign country uses an action response but the adult does not, this may partially explain why children become more fluent than adults. An intriguing question then is, how do adults compare with children when both apply action responses in controlled situations? In previous research (Asher, 1964; 1965; 1966; Kunihira & Asher, 1965) with a technique called the "Leaming Strategy of the Total Physical Response," it has been demonstrated that when adults team listening comprehension of either Russian or Japanese there is a highly signifieant difference in retention (p < .0005) if the adults are in action rather than passively writing English translations. For short-term experiments, it was found that dramatie differences in nnderstanding complex utterances in a foreign language are associated with action of Ss during retention tests, but not necessarily during training. For example, there were no significant differences when, in training, adult Ss listened to Russian commands and acted along with a model or merely sat and observed a model act. However, listening comprehension scores were vastly superior at p < .0005 when eacb S individualty listened to Russian commands and physically acted during retention tests, rather than sitting down, listening to Russian commands, and writing English translations. In the study to be described next, the listening comprehension for various complexities of Russian utterances was compared for children whose average ages were 8 years-1 month, 9 years-9 months, and 13 years-8 months. Also inctuded in the comparison were college adults between the ages of 18 and 2 1 . The ehitdren and adutts had identical training, and action responses were required of each S during retention tests. METHOD
.
:
Subfects The children (N = 96) were drawn from the second, fourth, and eighth grades of a public school^ in San Jose, Catifornia. The adults (N ~ 37) were college students recruited from xjndergraduate general psychology courses at San Jose State College. None of the children or adults had prior training or exposiire to Russian. The Ss from the 8-yearolds to adults were divided into two groups at each age level. The two groups, calted an "act-act group" and an "observe-act group," will be ^ Appreciation is expressed to Mr. William M. Phelps and the school personnel of the Blackford School in San Jose, California. 1220
JAMES J. ASHER AND BEN S. PRICE
described next. The average IQ for the cbildren in each group as measured by the Califomia Test of Mental Maturity was 115 for tbe 8-year-olds, 113 for the 10-year-olds, and 114 for the 14-year-old children. Procedure The act-act group.—In Unit I of the training, two Ss sat on eacb side of E and were instructed that a voice on tape would utter a command in Russian. When tbe Ss beard tbe Russian utterance, tbey were to do exactly what the instructor did. In the 3 minutes of training, Ss and tbe adult model pbysically responded to the following Russian commands: stand, sit, walk, stop, tum, squat, and run. Each utterance was presented ten times in a sequence whicb was varied so tbat Ss did not simply memorize a fixed pattem of bebavior. Tben a retention test was individually administered, followed by Unit II of training for nine minutes in which the Russian was increased to this complexity: Walk to the door; walk to the window; walk to the chair; walk to the table; run to the door; run to the window; run to tbe chair; run to tbe table. The session ended witb a retention test, and 24 bours later Ss returned for another retention test and Unit III of training. For 6 minutes, Ss and the model acted in response to Russian commands, such as: Pick up the pencil; put down the pencil; pick up the book; put down tbe book; pick up tbe paper; put down the paper; pick up the paper and pencil; put down the pencil, book, and paper. In tbe retention test, S physically responded to Russian utterances which tbey had heard in training, but tbey also heard novel Russian commands. Novelty was defined as the recombination of elements into sentences never experienced in training, as far example: Run to the table, and pick up the paper; pick up the pencil, and walk to the window; run to the chair, and put down the book. Forty-eigbt bours later, Ss returned for another retention test and tlien 732 minutes of Unit IV training, which consisted of the recombination and expansion of learned patterns, such as: Pick up the paper and pencil, and put them on the chair; run to tbe table, put down tbe paper, and sit on the cbair; walk to tbe door, pick up the pencil, put it on the table, and sit on the chair. At the end of tbis training there was anotber retention test, and 2 weeks later Ss retumed for a final retention test. The total amount of time in small-group training was 25!2 minutes as compared witb almost an equal amount of time in individual retention tests of 19 minutes. Observe-act group.—^The conditions were exactly the same as those for the act-act group, except that these Ss were instructed to sit silently in their chairs, listen to the Russian commands, and imagine tbat they 1221
CHILD DEVELOPMENT
were acting along with tbe model. Only the model perfonned while the Ss observed his behavior. However, during the retention tests each S listened individualty to commands in Russian and showed his understanding by acting (e.g., if the command in Russian was, "Walk to the chair and sit dovra," S indicated comprehension by physically walking to a chair and sitting down). RESULTS
We expected to find that the children and adults in the act-act situation would perform simitarty to the Ss in the observe-act situation. The reason is that prior research has shown that physical action during retention tests facilitated leaming. Since Ss in the act-act group and observe-act group all were in physical action during the recall tests, we expected no difference between treatments. The observe-act condition may then be viewed as a replication, in a sense, of the act-act; and, therefore, if the performance pattern is similar, this is an indieation of reliability. Indeed, Figures 1-5 show that children and adults in the act-act treatment perform quite like the children and adults in the observe-act treatment. The difference in retention of the Russian was between age groups. Adults performed near the maximum possible score in comprehension of
OBSERVE-AC!
ACT-ACT 1009080. O (/I
o> ^<
70. 6050. 403020-
j j
a.
100_ 10
14
YEARS YEARS YEARS OLD OLD OLD
ADULTS
8
10
YEARS
YEARS
14 YEARS
OLD
OLD
OLD
FIG. 1.—Retention of single Russian words.
1222
ADULTS
JAMES J. ASHER AND BEN S. PRICE
ACT-ACT
OBSERVE-ACT
IOO 90.
o
8 YEARS OLD
10 YEARS OLD
14 YEARS OLD
ADULTS
FIG. 2.—Retention of short Russian words.
OBSERVE-ACT
ACT-ACT
!00 - 01 LJ U-
90.
r o
80.
o
70.
^ CO
60. 50
Z
LJ
X
40. iO.
PERCEr
20. 10. 0
10 YEARS OLD
YEARS OLD
14 YEARS OLD
ADULTS YEARS OLD
10 YEARS OLD
14 ADULTS YEARS OLD
FIG. 3.—Retention of long Russian words.
1223
CHILD DEVELOPMENT
ACT-ACT
OBSERVE-ACT
100 O CO
o: o o 1x8 CO
u, O -s
UJ
8 10 YEARS YEARS OLD OLD
YEARS YEARS OLD OLD
14 ADULTS YEARS OLD
FIG. 4.—Retention of novel Russian words.
OBSERVE-ACT 100 O
o
IX O
%
50.
< a:
UJ
o cc UJ
a.
8 10 YEARS YEARS OLD OLD
14 YEARS OLD
8 10 14 ADULTS YEARS YEARS YEARS OLD OLD OLD FIG. 5.—Retention as measured by the total score from all retention tests.
JAMES J. ASHER AND BEN S. PRICE
Russian while the 8-year-old children were tbe lowest of all groups tested. Intermediate between adults and the 8-ycar-olds were the 10- and 14-yearold cbildren. Tbis generalization beld for single, short, long, or novel Russian utterances. A statistical analysis was obtained by applying two-tailed t tests. The results, presented in Tables 1 and 2, show tbat the adults, on the average TABLE 1 MEANS AND STANDARD DEVIATIONS FOR EACH AGE GROUP 2d GRADE (JV = 16) RETENTION TESTS
x:
SD
4th GRADE (N = 16)
X
8th GRADE
{N = 16)
SD
SD
ADULTS
iN = 18) X
SD
Act-Act Group Single Short Long Novel Total
22.56 41.43 66.68 34.87 146.31
7.58 17.86 33.86 17.61 57.73
28.93 54.25 95.31 49.93 191.50
9.09 16.68 27.76 16,17 51.56
32.68 59.25 98.43 51.68 208.00
7.54 21.89 37.13 23.47 67.34
44.72 2.10 87.22 5.47 173.33 21.00 103.11 15,81 323.61 27.07
Observe-Act Group Single Short Long Novel Total
21.81 40.81 79.12 41.50 154.93
7.16 16.29 28.86 14.46 49.48
29.81 46.31 85.25 41.93 173.75
10.05 21.35 35.67 19.31 64.57
32.50 64.81 106.31 56.68 220.87
9.66 21.74 46.05 29.44 74.34
43.05 3,29 86.42 3.24 172.26 19.36 101.68 14.56 318.26 23.14
and for any level of language complexity, dramatically excelled the cbildren of any age group tested. Every t between the adults and any group of childien was bigbly significant beyond the .0005 level of confidence. Cenerally, the 14-ycar-old students were not significantly different in retention when compared witb the 10-year-old students. However the older children, the 14-year-olds, tended to be significantly better than the youngest children, the 8-year-olds. The 10-year-old children excelled the 8-year-old children only in the act-act condition. DISCUSSION
Tbis study suggests that when adults leam a second language under the same conditions as children, the adults are superior. This generalization sbould be limited, at this time, to listening fluency. Future studies may show that cbildren have an advantage in fidelity of sound production. The comparison of college adults to children may be blurred somewhat because of a selectivity factor for the adults. The college students at San Jose State College are selected from among the top one-third of bigb school graduates in Califomia. Tberefore, we would expect above1225
CHILD DEVELOPMENT 1/1 in in I/) O
ooooo OOOQO ooooo
IO IO >O IO lo
ooooo ooooo ooooo
in t~-ooooo
3 O
O m
n O Om o
oooop ooooo ooooo CQ u w
n in
a:
1226
O in in lo IO
ooooo ooooo
JAMES J. ASHER AND BEN S. PRICE
average mental ability for tbe adults. However, it has been sbown by Pimsleur (1966) and others that general mental abihty is a hgbt-weigbt variable in second language leaming, accounting for less than 20 per cent of tbe variance. The second finding was that among children, the older child, the 10- or 14-year-old, tended to be significantly better tban tbe 8-year-old in his understanding of spoken Russian. Sbort attention span has been suggested as an explanation for the poor performance of the 8-year-old in understanding tbe Russian commands. When the Russian utterance was long and involved ("Pick up the pencil, walk to the chair, put down the pencil, and run to the window"), one might expect the 8-year-old child to have difficulty even if the utterance were spoken in English. Short attention span seems plausible for complex Russian commands, but this explanation does not account for the relatively low scores by tbe 8-year-oId for single Russian words as "run," "walk," and "sit." REFERENCES
Asher, J. J. Toward a neo-field theory of behavior. Journal of humanistic Psychology, 1964, 4, 85-94. Asher, J. J. The strategy of the total physical response: an application to leaming Russian. International Review of applied Linguistics, 1965, 3, 291-300. Asher, J. J. The leaming strategy of the total physical response: a review. Modern Language Journal, 1966, 50, 79-84. Kuniliira, S., & Asher, J. J. The strategy of the total physical response: an application to leaming Japanese. International Review of applied Linguistics, 1965, 3, 277-289. Pimsleur, P. In Albert Valdman (Ed.), Trends in hnguage teaching. New York: McGraw-HiU, 1966. Pp. 175-214.
1227
