Brian K. Weiler C. Melanie Schuele INTRODUCTION METHODS ...
An Exploration of Past Tense Marking and Lexical Aspect Brian K. Weiler w C. Melanie Schuele Vanderbilt University School of Medicine
Past tense verb marking emerges during the preschool years. By the age of five, children with typical language skills mark finiteness on regular and irregular past tense at levels approaching mastery (Rice & Wexler, 1996; Rice, Wexler, Marquis & Hershberger, 2000). Prior to mastery, young children sometimes produce bare verb stems (e.g.,play) in contexts where inflected forms (e.g., played) are obligatory in the adult language. Potential influences on partial past tense marking previously investigated include lexical frequency, phonological composition, and neighborhood structure (Marchman, Welteck, & Wesimer, 1999), phonotactic probability (Leonard, Davis, & Deevy, 2007), sentence position (Dalal & Loeb, 2005) and, of present interest, lexical aspect (Johnson & Fey, 2006; Johnson & Morris, 2007). Lexical aspect relates to the temporal properties of events individual verbs describe (Li & Shirai, 2000).
Fifty-two preschool children aged 37 to 61 months participated in the current study. These children were identified from a broader pool of 100 children with typical language skills (see Table 2). Participants were selected for variable past tense marking across sixteen elicited past tense verb targets (i.e., at least one marked and one omitted in obligatory contexts). All children produced word final /d/ and /t/ on a phonological probe. The children were recruited from four preschools. Demographic information received from 83% of the final participant pool indicated that 98% of the children had at least one parent with a college education and that 98% were Caucasian. All participants were monolingual speakers of mainstream English dialect.
whether an event takes Hme to occur [+ duraHve] or occurs instantaneously [-‐ duraHve].
dynamic
whether an event involves moHon [+ dynamic] or is a state [ -‐ dynamic]. whether an event has a specified endpoint [+ telic] or an undefined endpoint [ -‐ telic].
telic
Table 2. LEXICAL ASPECT CATEGORIES Category State AcHvity Accomplishment Achievement
Temporal proper2es [+ duraHve] [-‐ dynamic] [-‐ telic]
[+ duraHve] [+ dynamic] [-‐ telic]
[+ duraHve] [+ dynamic] [+ telic]
Example u7erances I wanted it.
She rode the horse.
He 1ed his shoes.
Age in Months
48.94
7.01
PPVT-‐III SS
110.48
11.31
PLS-‐4 Total SS
116.61
13.07
METHODS The study involved secondary analysis of data collected for an investigation of complex syntax in typical language learners (Schuele, 2006). Child responses to 16 verb targets from the past tense probe of the Rice/Wexler Test of Early Grammatical Impairment (TEGI; Rice & Wexler, 2001) were analyzed. The 16 TEGI verb targets were selected for their lexical aspect categorization as either Accomplishment or Achievement using Shirai and Anderson’s (1995) coding scheme. Table 3. VERB TARGETS BY CATEGORY Achievement kick, jump, catch, blow, li;, give
[-‐ duraHve] [+ dynamic] [+ telic]
She blew out the candle.
Brown (1973) observed that young children first mark past tense for Achievements like fell, dropped, slipped, crashed, and broke “which name events of such brief duration that the event is almost certain to have ended before one can speak” (p. 334). Follow-up analyses of Brown’s Adam and Eve language samples, at ages 2;3 and 1;6 respectively, show that 94-100% of past tense marking was for Achievements. Bloom, Lifter, and Hafitz (1980) likewise reported that young children’s use of past tense overwhelmingly occurred when referencing completed events of instantaneous duration. Johnson and Fey (2006) found that two-year-olds with typical language imitate past tense morphology with higher accuracy for Accomplishments than Activities. The temporal distinction between Accomplishments and Activities relates to telicity. Leonard et al. (2007) found that, in an elicitation probe, three-year-olds with typical language mark regular past tense with greater accuracy for Achievements than Activities. The temporal distinction between Achievements and Activities relates to telicity and duration. It is unclear whether the facilitative advantage in marking past tense for Activities when compared to Achievements was related more to the property of telicity or duration. To answer this question, a comparison of past tense marking for Achievements versus Accomplishments, both of which are + telic, is required. ♦ The purpose of this study was to explore the influence of the durative property on past tense marking among typical preschoolers who have not mastered past tense morphology. ♦ Accuracies in marking past tense where inflected forms were obligatory were compared across Achievements and Accomplishments.
90
Accomplishment clean, climb, made, write, paint, pick, eat, brush, plant, 1e
Task: The TEGI past tense probe is comprised of picture sets used to elicit regular and irregular past tense verbs. For each item, an examiner showed the child two pictures, one of an action in progress and the other of the same action completed. The examiner described the first picture using the target verb in a present progressive context (e.g., Here the boy is cleaning her room.). Then, the child was asked to describe the second, completed action picture using the prompt, Now he/she is done. Tell me what he/she did. The child was prompted to respond with a complete sentence, including a subject. Data Analysis: For the purposes of this study, only those responses containing the target verb in an obligatory context for past tense were coded for analysis (see Table 4). Correctly marked regular and irregular verbs were given credit. Verbs taking irregular past tense forms (e.g., blow) that were marked with regular past tense overgeneralizations of either the bare stem (e.g., blowed) or the past tense form itself (e.g., blewed) were also given credit. Kuczac (1977) found that such past tense overgeneralization errors are produced by children who have already demonstrated stable control of the regular past tense rule. Responses containing the bare stem of the target verb in an obligatory context for past tense (e.g., She clean her room.) were not given credit. Interscorer reliability was 90%. Table 4. CREDITED CHILD RESPONSES Correct regular or irregular past
He kicked. She made.
Over-‐regularizaHon of irregular past
She maded.
Regular past over-‐marked
He jumped-ed
Irregular bare stem over-‐marked
She catched-ed
Irregular past over-‐marked
She caughted-ed
Achievement
80 70
blow jump
60
Percent Marked 91.7 90.0
50
catch
87.5
40
kick
81.6
30
give
72.5
li,*
20
16.1
10 0
* = alveolar bare stem verb Achievement
Accomplishment
Accomplishment
Percent Marked
made
92.3
1e
91.7
pick
90.7
climb
87.8
clean
80.5
brush
80.0
eat*
76.6
paint*
62.2
plant*
57.6
write*
51.4
Main Finding Total past tense marking accuracies were similar across Lexical Aspect categories differing on the durative property. (n = 52 children) Achievement Verbs: 75.55% (173/229) Accomplishment Verbs: 76.94% (307/399) Figure 2. AGE 3 ACCURACY
Figure 3. AGE 4 ACCURACY
100
100
90
90
80 70 60 50 40 30 20
80 70 60 50 40 30 20
10
10
0
0
Achievement
Accomplishment
Achievement
Accomplishment
76.56% 76.19% (n = 28 children)
73.68% 76.54% (n = 23 children)
Figure 4. NON-ALVEOLAR PAST TENSE ACCURACY 100 90
Percent Marked for Past Tense
duraHve
SD
Table 5. PERCENT ACCURACIES BY VERB
100
Table 3. PARTICIPANT CHARACTERISTICS Mean
RESULTS
Figure1. OVERALL PAST TENSE ACCURACY
Percent Marked for Past Tense
Table 1. TEMPORAL PROPERTIES DEFINED
RESULTS
Percent Marked for Past Tense
PARTICIPANTS
Percent Marked for Past Tense
INTRODUCTION
DISCUSSION The findings of this study suggest that the instantaneous property of Achievements does not provide a facilitative advantage over the durative property of Accomplishments in past tense marking among typical preschool children who have not mastered past tense morphology. Similar overall accuracies in past tense marking were found across the two lexical aspect categories under investigation. This similarity was also found at individual age levels and when alveolar bare stem verbs, found presently and in previous studies (e.g., Marchman, 1997) to be more susceptible to zero-marking for past tense, were excluded. Prior work (e.g., Brown 1973) indicates that early emerging spontaneous past tense productions frequently occur with Achievements. This is speculated to be due to the prototypical way in which verbs describing Achievements characterize visible events completed rather instantly. The present results suggest that the temporal property of Achievements implicated in facilitating early emerging past tense productions is not facilitative of later preschool-aged past tense marking above and beyond the previously documented effects of telicity (Leonard et al. 2007). Considering the current findings alongside prior investigations of lexical aspect and past tense marking, it seems that typical preschoolers are better able to exploit their past tense morphology skills with verbs that describe an event with a specified endpoint but are relatively uninfluenced by whether that endpoint was reached instantaneously (Achievement) or over time (Accomplishment).
80 70 60 50 40
ACKNOWLEDGEMENTS
30 20 10 0
Achievement
Accomplishment
Post-hoc Finding Total past tense marking accuracies on non-alveolar bare stem verbs were similar across Lexical Aspect categories differing on the durative property. (n = 26 children) Achievement Verbs: 74.29% (78/105) Accomplishment Verbs: 77.42% (96/124)
Data collection was supported by NIH/NIDCD (PI: Schuele). Completion of this study and preparation of this poster were supported by a Preparation of Leadership Personnel grant (H325D080075; PI: Schuele), U.S. Department of Education. LIST OF CITED REFERENCES AVAILABLE ON REQUEST www.mc.vanderbilt.edu/LANGUAGELAB Poster Presented at the Annual ASHA ConvenHon, San Diego, CA November 2011
Past tense verb marking emerges during the preschool years. By the age of five, children with typical language skills mark finiteness on regular and irregular past tense at levels approaching mastery (Rice & Wexler, 1996; Rice, Wexler, Marquis & Hershberger, 2000). Prior to mastery, young children sometimes produce bare verb stems (e.g.,play) in contexts where inflected forms (e.g., played) are obligatory in the adult language. Potential influences on partial past tense marking previously investigated include lexical frequency, phonological composition, and neighborhood structure (Marchman, Welteck, & Wesimer, 1999), phonotactic probability (Leonard, Davis, & Deevy, 2007), sentence position (Dalal & Loeb, 2005) and, of present interest, lexical aspect (Johnson & Fey, 2006; Johnson & Morris, 2007). Lexical aspect relates to the temporal properties of events individual verbs describe (Li & Shirai, 2000).
Fifty-two preschool children aged 37 to 61 months participated in the current study. These children were identified from a broader pool of 100 children with typical language skills (see Table 2). Participants were selected for variable past tense marking across sixteen elicited past tense verb targets (i.e., at least one marked and one omitted in obligatory contexts). All children produced word final /d/ and /t/ on a phonological probe. The children were recruited from four preschools. Demographic information received from 83% of the final participant pool indicated that 98% of the children had at least one parent with a college education and that 98% were Caucasian. All participants were monolingual speakers of mainstream English dialect.
whether an event takes Hme to occur [+ duraHve] or occurs instantaneously [-‐ duraHve].
dynamic
whether an event involves moHon [+ dynamic] or is a state [ -‐ dynamic]. whether an event has a specified endpoint [+ telic] or an undefined endpoint [ -‐ telic].
telic
Table 2. LEXICAL ASPECT CATEGORIES Category State AcHvity Accomplishment Achievement
Temporal proper2es [+ duraHve] [-‐ dynamic] [-‐ telic]
[+ duraHve] [+ dynamic] [-‐ telic]
[+ duraHve] [+ dynamic] [+ telic]
Example u7erances I wanted it.
She rode the horse.
He 1ed his shoes.
Age in Months
48.94
7.01
PPVT-‐III SS
110.48
11.31
PLS-‐4 Total SS
116.61
13.07
METHODS The study involved secondary analysis of data collected for an investigation of complex syntax in typical language learners (Schuele, 2006). Child responses to 16 verb targets from the past tense probe of the Rice/Wexler Test of Early Grammatical Impairment (TEGI; Rice & Wexler, 2001) were analyzed. The 16 TEGI verb targets were selected for their lexical aspect categorization as either Accomplishment or Achievement using Shirai and Anderson’s (1995) coding scheme. Table 3. VERB TARGETS BY CATEGORY Achievement kick, jump, catch, blow, li;, give
[-‐ duraHve] [+ dynamic] [+ telic]
She blew out the candle.
Brown (1973) observed that young children first mark past tense for Achievements like fell, dropped, slipped, crashed, and broke “which name events of such brief duration that the event is almost certain to have ended before one can speak” (p. 334). Follow-up analyses of Brown’s Adam and Eve language samples, at ages 2;3 and 1;6 respectively, show that 94-100% of past tense marking was for Achievements. Bloom, Lifter, and Hafitz (1980) likewise reported that young children’s use of past tense overwhelmingly occurred when referencing completed events of instantaneous duration. Johnson and Fey (2006) found that two-year-olds with typical language imitate past tense morphology with higher accuracy for Accomplishments than Activities. The temporal distinction between Accomplishments and Activities relates to telicity. Leonard et al. (2007) found that, in an elicitation probe, three-year-olds with typical language mark regular past tense with greater accuracy for Achievements than Activities. The temporal distinction between Achievements and Activities relates to telicity and duration. It is unclear whether the facilitative advantage in marking past tense for Activities when compared to Achievements was related more to the property of telicity or duration. To answer this question, a comparison of past tense marking for Achievements versus Accomplishments, both of which are + telic, is required. ♦ The purpose of this study was to explore the influence of the durative property on past tense marking among typical preschoolers who have not mastered past tense morphology. ♦ Accuracies in marking past tense where inflected forms were obligatory were compared across Achievements and Accomplishments.
90
Accomplishment clean, climb, made, write, paint, pick, eat, brush, plant, 1e
Task: The TEGI past tense probe is comprised of picture sets used to elicit regular and irregular past tense verbs. For each item, an examiner showed the child two pictures, one of an action in progress and the other of the same action completed. The examiner described the first picture using the target verb in a present progressive context (e.g., Here the boy is cleaning her room.). Then, the child was asked to describe the second, completed action picture using the prompt, Now he/she is done. Tell me what he/she did. The child was prompted to respond with a complete sentence, including a subject. Data Analysis: For the purposes of this study, only those responses containing the target verb in an obligatory context for past tense were coded for analysis (see Table 4). Correctly marked regular and irregular verbs were given credit. Verbs taking irregular past tense forms (e.g., blow) that were marked with regular past tense overgeneralizations of either the bare stem (e.g., blowed) or the past tense form itself (e.g., blewed) were also given credit. Kuczac (1977) found that such past tense overgeneralization errors are produced by children who have already demonstrated stable control of the regular past tense rule. Responses containing the bare stem of the target verb in an obligatory context for past tense (e.g., She clean her room.) were not given credit. Interscorer reliability was 90%. Table 4. CREDITED CHILD RESPONSES Correct regular or irregular past
He kicked. She made.
Over-‐regularizaHon of irregular past
She maded.
Regular past over-‐marked
He jumped-ed
Irregular bare stem over-‐marked
She catched-ed
Irregular past over-‐marked
She caughted-ed
Achievement
80 70
blow jump
60
Percent Marked 91.7 90.0
50
catch
87.5
40
kick
81.6
30
give
72.5
li,*
20
16.1
10 0
* = alveolar bare stem verb Achievement
Accomplishment
Accomplishment
Percent Marked
made
92.3
1e
91.7
pick
90.7
climb
87.8
clean
80.5
brush
80.0
eat*
76.6
paint*
62.2
plant*
57.6
write*
51.4
Main Finding Total past tense marking accuracies were similar across Lexical Aspect categories differing on the durative property. (n = 52 children) Achievement Verbs: 75.55% (173/229) Accomplishment Verbs: 76.94% (307/399) Figure 2. AGE 3 ACCURACY
Figure 3. AGE 4 ACCURACY
100
100
90
90
80 70 60 50 40 30 20
80 70 60 50 40 30 20
10
10
0
0
Achievement
Accomplishment
Achievement
Accomplishment
76.56% 76.19% (n = 28 children)
73.68% 76.54% (n = 23 children)
Figure 4. NON-ALVEOLAR PAST TENSE ACCURACY 100 90
Percent Marked for Past Tense
duraHve
SD
Table 5. PERCENT ACCURACIES BY VERB
100
Table 3. PARTICIPANT CHARACTERISTICS Mean
RESULTS
Figure1. OVERALL PAST TENSE ACCURACY
Percent Marked for Past Tense
Table 1. TEMPORAL PROPERTIES DEFINED
RESULTS
Percent Marked for Past Tense
PARTICIPANTS
Percent Marked for Past Tense
INTRODUCTION
DISCUSSION The findings of this study suggest that the instantaneous property of Achievements does not provide a facilitative advantage over the durative property of Accomplishments in past tense marking among typical preschool children who have not mastered past tense morphology. Similar overall accuracies in past tense marking were found across the two lexical aspect categories under investigation. This similarity was also found at individual age levels and when alveolar bare stem verbs, found presently and in previous studies (e.g., Marchman, 1997) to be more susceptible to zero-marking for past tense, were excluded. Prior work (e.g., Brown 1973) indicates that early emerging spontaneous past tense productions frequently occur with Achievements. This is speculated to be due to the prototypical way in which verbs describing Achievements characterize visible events completed rather instantly. The present results suggest that the temporal property of Achievements implicated in facilitating early emerging past tense productions is not facilitative of later preschool-aged past tense marking above and beyond the previously documented effects of telicity (Leonard et al. 2007). Considering the current findings alongside prior investigations of lexical aspect and past tense marking, it seems that typical preschoolers are better able to exploit their past tense morphology skills with verbs that describe an event with a specified endpoint but are relatively uninfluenced by whether that endpoint was reached instantaneously (Achievement) or over time (Accomplishment).
80 70 60 50 40
ACKNOWLEDGEMENTS
30 20 10 0
Achievement
Accomplishment
Post-hoc Finding Total past tense marking accuracies on non-alveolar bare stem verbs were similar across Lexical Aspect categories differing on the durative property. (n = 26 children) Achievement Verbs: 74.29% (78/105) Accomplishment Verbs: 77.42% (96/124)
Data collection was supported by NIH/NIDCD (PI: Schuele). Completion of this study and preparation of this poster were supported by a Preparation of Leadership Personnel grant (H325D080075; PI: Schuele), U.S. Department of Education. LIST OF CITED REFERENCES AVAILABLE ON REQUEST www.mc.vanderbilt.edu/LANGUAGELAB Poster Presented at the Annual ASHA ConvenHon, San Diego, CA November 2011
