SnedLab
VE
Onthepsychologicalrealityoflinguis3ceventstructures
RI
TAS
H
A R V AR D
Jayden Ziegler ([email protected]), Jesse Snedeker • Harvard Psychology
Lab for Developmental Studies
Departmenet of Psychology • FAS
1.Introduc3on
3.Valida3on
•  Structural priming: tendency to repeat aspects of structure across sentences (Bock, 1986)
?  No cross-structural priming in general? PreposiKonal mismatch? Reducible to animacy?
•  SKmuli:
EXP.
N
PRIMES
TARGETS
4♫
52
benefacKves daKves
=parallel structure
5♬
52
provide-withs daKves
=parallel structure
6
172
provide-withs locaKves
=parallel structure?
-----------------
7 (≈2)
172
locaKves
locaKves
=parallel structure
+ animacy mismatch (animate locaKon in primes, inanimate in targets)
8 (≈3)
172
locaKves
daKves
=parallel structure?
+animacy match (animate locaKon, animate recipient)
•  Most priming reducible to parallel surface syntax (though cf. Chang et al., 2003; for review, see Pickering &
Ferreira, 2008)
Rappaport Hovav & Levin, 2008; for review, see Levin & Rappaport Hovav, 2005)
2.Methods/Results
•  AnimaKon descripKon task on MTurk (N=1,068; age range=18-69)
•  SKmuli:
EXP.
1
2♩
N
52
52
PRIMES
daKves
locaKves
TARGETS
daKves
locaKves
=parallel structure
=parallel structure
3*
172
locaKves
daKves
=parallel structure?
Proportion of TF responses
?  Can semanKc structure also be primed? If so, at what level?
•  Test case:
DATIVES
and LOCATIVES
boy brings camel keys
—
=ditransiKve
—
girl loads van with boxes =causaKve+with
boy brings keys to camel
girl loads boxes in van
=to-variant/caused-moKon
•  At stake: degree of abstracKon/generalizaKon of semanKc structure
1.  Caused-moKon primes to-variant à parallel structure (Jackendoff, 1983; Goldberg, 1995)
2.  Lack of priming à appropriate grain size at level of individual event structures (Pinker, 1989;
Proportion of TF responses
Exp. 1: dative−dative
***
1.00
Exp. 2: locative−locative
***
Syntax?
***
***
1.00
Syntax?
Exp. 6
Not
syntax
ns.
Exp. 7
Exp. 8
**
*
NotP
0.75
0.50
0.25
0.00
GF
TF
GF
TF
GF
TF
GF
TF
GF
TF
1989; Pappert & Pechmann, 2013; ♬Hare & Goldberg, 1999; Salamoura & Williams, 2007; Cho-Reyes et al., 2016
Exp. 3: locative−dative
4.Discussion/Conclusions
ns.
² Event structures are psychologically real
1.  Priming of individual event structures, independent of animacy (Bock et al., 1992; CarminaK et al.,
0.75
0.50
2008; Cho-Reyes et al., 2016)
2.  No strong evidence for higher-level abstracKons (though plausibly sKll there)
0.25
0.00
GF
TF
GF
TF
Prime Type
♩Chang
Not
syntax,
butP?
Exp. 5
Prime Type
♫Bock,
Exp. 4
GF
TF
•  Parallel animacy facilitates priming (Gámez & Vasilyeva, 2015)
1.  Animacy changes inherent nature of event structural representaKons?
2.  Animacy as truly orthogonal dimension, and priming instead sensiKve to role filler
properKes? (Bock et al., 1992; Levin & Rappaport Hovav, 2005; cf. Branigan et al., 2008)
et al., 2003; *Also run in reverse direcKon (N=172), with similar results
References: [•] Bock, K. (1986). SyntacKc persistence in language producKon. Cogni&ve Psychology, 18, 355-387. [•] Bock, K. (1989). Closed-class immanence in sentence producKon. Cogni&on, 31(2), 163-186. [•] Bock, K., Loebell, H., & Morey, R. (1992). From conceptual roles to structural relaKons: Bridging the syntacKc cleY. Psychological Review, 99, 150-171. [•] Branigan, H. P., Pickering, M. J., & Tanaka, M. (2008). ContribuKons of animacy to
grammaKcal funcKon assignment and word order during producKon. Lingua, 118(2), 172-189. [•] CarminaK, M. N., van Gompel, R. P. G., Scheepers, C., & Arai, M. (2008). SyntacKc priming in comprehension: The role of argument order and animacy. Journal of Experimental Psychology: Learning, Memory, and Cogni&on, 34(5), 1098-1110. [•] Chang, F., Bock, K., & Goldberg, A. E. (2003). Can themaKc roles leave traces of their places? Cogni&on, 90,
29-49. [•] Cho-Reyes, S., Mack, J. E., & Thompson, C. K. (2016, in press). GrammaKcal encoding and learning in agrammaKc aphasia: Evidence from structural priming. Journal of Memory and Language. [•] Gámez, P. B., & Vasilyeva, M. (2015). Exploring interacKons between semanKc and syntacKc processes: The role of animacy in syntacKc priming. Journal of Experimental Child Psychology, 138, 15-30. [•] Goldberg, A. E. (1995). Construc&ons: A
construc&on grammar approach to argument structure. Chicago: University of Chicago Press. [•] Hare, M. L., & Goldberg, A. E. (1999). Structural priming: Purely syntacKc? In M. Hahn & S. C. Stones (Eds.), Proceedings of the 21st Annual Mee&ng of the Cogni&ve Science Society (pp. 208-211). Mahwah: Erlbaum. [•] Jackendoff, R. S. (1983). Seman&cs and cogni&on. Cambridge: MIT Press. [•] Levin, B., & Rappaport Hovav, M. (2005). Argument
realizaKon. Cambridge: Cambridge University Press. [•] Pappert, S., & Pechmann, T. (2013). BidirecKonal structural priming across alternaKons: Evidence from the generaKon of daKve and benefacKve alternaKon structures in German. Language and Cogni&ve Processes, 28, 1303-1322. [•] Pickering, M. J., & Ferreira, V. S. (2008). Structural priming: A criKcal review. Psychological Bulle&n, 134(3), 427-459. [•] Pinker, S. (1989). Learnability and cogni&on:
The acquisi&on of argument structure. Cambridge: MIT Press. [•] Rappaport Hovav, M., & Levin, B. (2008). The English daKve alternaKon: The case for verb sensiKvity. Journal of Linguis&cs, 44(1), 129-167. [•] Salamoura, A, & Williams, J. N. (2007). Processing verb argument structure across languages: Evidence for shared representaKons in the bilingual lexicon. Applied Psycholinguis&cs, 28, 627-660.
Acknowledgments: This work was supported in part by the Norman Henry Anderson Graduate Psychology Fund from Harvard GSAS. Task animaKons/images created by Brianne Gallagher and Anthony Yacovone. Special thanks to Steve Pinker, Harvard LDS, Harvard Psychology, Anthony Yacovone, and audiences at CUNY 2016 for criKcal discussion of this work. Thanks also to Jared Hawn for help with coding.
CogSci 2016, August 10-13, 2016 • Philly, PA