Li and Abarbanell (2019) showed that English-learning 4- to 5-year-olds' relative ease of learning an egocentric versus an allocentric spatial rule in a two-table hidden-object paradigm shifts with task context (table distance, type of 180-degree turn, and array axis), undermining the claim of a phylogenetically inherited allocentric bias. The authors themselves left open whether this flexibility characterizes children acquiring geocentric languages such as Tseltal or Hai||om, and whether performance in such rule-guessing paradigms reflects genuinely cognitive encoding preferences or Gricean pragmatic strategies shaped by linguistic conventions. No published study has tested geocentric-language preschoolers on this paradigm; work on older Tseltal children, Japanese preschoolers, and adults in geocentric communities addresses neighboring questions but leaves the question open.

Background

A spatial frame of reference (FoR) is a coordinate system used to specify where things are: an egocentric (body-defined) FoR locates objects relative to the speaker's own body axes (e.g., "to my left"), while an allocentric or geocentric (environment-defined) FoR locates objects relative to fixed environmental anchors such as terrain slope or cardinal directions (e.g., "uphill", "north of the cup"). Languages differ strikingly in which FoR their speakers habitually use for small-scale space: English favors egocentric terms, while in some languages, such as Tseltal (a Mayan language of Chiapas, Mexico) and ≠Akoe Hai||om (a Khoisan language of Namibia), speakers preferentially use geocentric terms even for tabletop-scale descriptions. This variation fuels the linguistic-relativity debate over whether the language one learns shapes nonlinguistic spatial cognition. Haun, Rapold, Call, Janzen and Levinson (2006) asked whether there is a phylogenetically inherited FoR preference underlying language acquisition. In their hidden-object rule-guessing task, a child sees a reward hidden in one of two or three cups on a first table, walks a 180-degree arc to a second table, and must find the reward by inferring the experimenter's rule: an egocentric rule (e.g., "always the cup on your left") or an allocentric rule (e.g., "always the cup toward the same environmental side"). They reported that German-speaking 4-year-olds, like nonhuman great apes, performed better on the allocentric rule, whereas 8-year-olds and adults diverged by language (Dutch speakers favored egocentric; Hai||om speakers solved only the allocentric rule). They concluded that humans inherit an allocentric bias that speaking a language with egocentric terms later culturally overrides — a widely cited result. Li and Abarbanell (2019) re-examined this conclusion in four experiments with English-learning preschoolers. Replicating the original arrangement (Experiment 1) reproduced the allocentric advantage, but increasing the distance between tables and changing the turn so children no longer faced the first table (Experiments 2 and 4) equalized the two rules, and aligning the cups with the children's front-back axis rather than left-right axis (Experiment 3) even reversed the advantage. Because spatial updating (the automatic recomputation of object locations as one moves, which favors environmental coding when the two views share a line of sight) and shared-view structure differ across these arrangements, Li and Abarbanell argued the original allocentric advantage was an artifact of task constraints: young children's egocentric and allocentric representations are both readily available, and salience is context-sensitive. The discussion raised open issues it could not settle: whether geocentric-language-learning preschoolers show the same flexibility; whether adults' and older children's language-congruent rule guessing reflects genuine encoding preferences or Gricean pragmatic reasoning (inference about the experimenter's communicative intent, of the kind studied in pragmatics since Grice); why the words "left" and "right" are acquired years later than "front" and "back" (classic acquisition studies reviewed in the source) although rule learning in Experiment 4 showed no such axis asymmetry (the source speculates that the hypothesis space for rule learning may differ from that for word learning, and that left-right symmetry and the difficulty of non-egocentric left-right mapping play a role, but it does not test these proposals); whether still larger displacements would yield a clear egocentric advantage, as Rosati (2015) found in bonobos tested at larger table separations; and how the factors that modulate FoR salience interact during naturalistic spatial word learning. Related prior evidence includes Li, Abarbanell, Gleitman and Papafragou (2011), who found Tseltal-speaking adults were often better with egocentric than allocentric spatial relations, Shusterman and Li (2016) on English-learning 4-year-olds readily acquiring geocentric novel words, and Li and Abarbanell (2018) for the authors' broader framework.

Problem Statement

The task is to determine whether the task-context flexibility that Li and Abarbanell (2019) demonstrated in English-learning 4- to 5-year-olds is a language-general property of preschool spatial cognition — adjudicating between the inherited-allocentric-bias-with-cultural-override account (Haun et al., 2006) and the context-sensitive-flexibility account. The central question is (a): what pattern of flexibility between egocentric and allocentric rules is expected for 4- to 5-year-olds acquiring a geocentric language, what evidence supports that expectation, and what design would test it? The related open questions are: (a) Test 4- to 5-year-old children acquiring a geocentric language (e.g., Tseltal, Hai||om) on the Li–Abarbanell table setups (varying table distance, turn type, and cup-array axis): do they show the same pattern of flexibility between egocentric and allocentric rules as English-learning children, or does early exposure to a geocentric linguistic system interact with these manipulations? (b) Evaluate the source's untested Gricean account by disentangling genuinely cognitive encoding preferences from pragmatic strategies shaped by linguistic conventions: does language-congruent rule guessing in older children and adults from different linguistic backgrounds reflect how spatial relations are encoded, or how the experimenter's communicative intent is interpreted? (c) Explain why, in natural word learning, 'left' and 'right' are acquired much later than 'front' and 'back' although children perform equally well on left-right and front-back egocentric rules in the rule-learning paradigm. (d) Establish whether even greater table distances, or testing across entirely different rooms, produce a clear egocentric advantage on the left-right axis, as Rosati's (2015) findings with bonobos suggest. (e) Investigate whether these laboratory findings inform how children acquire frame-of-reference words in naturalistic learning — the source speculates that left-right word learning "may have a better outcome when the words are reintroduced across a longer distance, where it is difficult to track geocentric relations", and notes potential implications for how best to teach these meanings.

References

  1. Li, P., & Abarbanell, L. (2019). Alternative spin on phylogenetically inherited spatial reference frames. Cognition, 191, 103983. doi:10.1016/j.cognition.2019.05.020 https://doi.org/10.1016/j.cognition.2019.05.020
  2. Haun, D. B. M., Rapold, C. J., Call, J., Janzen, G., & Levinson, S. C. (2006). Cognitive cladistics and cultural override in Hominid spatial cognition. Proceedings of the National Academy of Sciences, 103(46), 17568–17573. doi:10.1073/pnas.0607999103 https://doi.org/10.1073/pnas.0607999103
  3. Haun, D. B. M., Rapold, C. J., Janzen, G., & Levinson, S. C. (2011). Plasticity of human spatial cognition: Spatial language and cognition covary across cultures. Cognition, 119(1), 70–80. doi:10.1016/j.cognition.2010.12.009 https://doi.org/10.1016/j.cognition.2010.12.009
  4. Li, P., Abarbanell, L., Gleitman, L., & Papafragou, A. (2011). Spatial reasoning in Tenejapan Mayans. Cognition, 120(1), 33–53. doi:10.1016/j.cognition.2011.02.012 https://doi.org/10.1016/j.cognition.2011.02.012
  5. Li, P., & Abarbanell, L. (2018). Competing perspectives on frames of reference in language and thought. Cognition, 170, 9–24. doi:10.1016/j.cognition.2017.09.005 https://doi.org/10.1016/j.cognition.2017.09.005
  6. Shusterman, A., & Li, P. (2016). Frames of reference in spatial language acquisition. Cognitive Psychology, 88, 115–161. doi:10.1016/j.cogpsych.2016.06.001 https://doi.org/10.1016/j.cogpsych.2016.06.001
  7. Rosati, A. G. (2015). Context influences spatial frames of reference in bonobos (Pan paniscus). Behaviour, 152(3–4), 375–406. doi:10.1163/1568539X-00003189 https://doi.org/10.1163/1568539X-00003189
  8. Abarbanell, L., & Li, P. (2021). Unraveling the contribution of left-right language on spatial perspective taking. Spatial Cognition and Computation, 21(1), 1–38. doi:10.1080/13875868.2020.1825442 https://doi.org/10.1080/13875868.2020.1825442
  9. Marghetis, T., McComsey, M., & Cooperrider, K. (2020). Space in hand and mind: Gesture and spatial frames of reference in bilingual Mexico. Cognitive Science, 44(12), e12920. doi:10.1111/cogs.12920 https://doi.org/10.1111/cogs.12920
  10. Pitt, B., Carstensen, A., Boni, I., Piantadosi, S. T., & Gibson, E. (2022). Different reference frames on different axes: Space and language in indigenous Amazonians. Science Advances, 8(47), eabp9814. doi:10.1126/sciadv.abp9814 https://doi.org/10.1126/sciadv.abp9814
  11. Pitt, B. (2025). One action, two reference frames: Compound cognitive maps of object location. Psychological Science, 36(11), 862–873. doi:10.1177/09567976251391172 https://doi.org/10.1177/09567976251391172
  12. Zheng, X., & Sugimura, S. (2024). Effects of landmark availability on Japanese speakers' preferences for spatial frames of reference. Japanese Psychological Research. doi:10.1111/jpr.12545 https://doi.org/10.1111/jpr.12545
  13. Stibbard-Hawkes, D. N. E., Abarbanell, L., Mabulla, I. A., Endeko, E. S., Legare, C. H., & Apicella, C. L. (2024). Foreign-language effects in cross-cultural behavioral research: Evidence from the Tanzanian Hadza. PNAS Nexus, 3(6), pgae218. doi:10.1093/pnasnexus/pgae218 https://doi.org/10.1093/pnasnexus/pgae218
  14. Xu, Q., Ya, R., Zhang, E., Li, J., Ah, R., & Li, M. (2023). Cardinal direction judgment based on the integration of spatial reference frames in different languages. Functions of Language, 30(3), 320–348. doi:10.1075/fol.22038.xu https://doi.org/10.1075/fol.22038.xu