Compression as a Universal Principle of Animal Behavior

Statistical Language and Speech Processing

et al. 2016

Self Cite

Abstract. The pioneering research of G. K. Zipf on the relationship between word frequency and other word features led to the formulation of various linguistic laws. Here we focus on a couple of them: the meaning-frequency law, i.e. the tendency of more frequent words to be more polysemous, and the law of abbreviation, i.e. the tendency of more frequent words to be shorter. Here we evaluate the robustness of these laws in contexts where they have not been explored yet to our knowledge. The recovery of the laws again in new conditions provides support for the hypothesis that they originate from abstract mechanisms.

Section: Introductionmentioning

confidence: 99%

Testing the Robustness of Laws of Polysemy and Brevity Versus Frequency

Hernández-Fernández

Fernández

Statistical Language and Speech Processing

et al. 2016

Self Cite

“…Analyses of the nonvocal surface behavioral repertoire of dolphins (7), the vocal repertoire of Formosan macaques (8), close-range calls of common marmosets (9), and social calls of four species of bats (10) reveal that they too conform to an inverse general relationship between magnitude (e.g., duration) and frequency of use. This common pattern provides evidence that compression-the informationtheoretic principle of minimizing the expected length of a codeis a general principle of animal (including human) behavior, reflecting selection for energetic efficiency (5,11).…”

mentioning

confidence: 97%

“…The power of such an approach is illustrated by recent studies of the generality of Zipf's law of abbreviation (6). This linguistic law, also commonly known as Zipf's law of brevity, states that more frequently used words tend to be shorter, and it has been found to hold true in all languages assessed to date (5). Analyses of the nonvocal surface behavioral repertoire of dolphins (7), the vocal repertoire of Formosan macaques (8), close-range calls of common marmosets (9), and social calls of four species of bats (10) reveal that they too conform to an inverse general relationship between magnitude (e.g., duration) and frequency of use.…”

mentioning

confidence: 99%

“…The existence of such principles, revealed by the occurrence of common statistical patterns, can shed light on the basic processes shaping biological systems (3). In recent years, exploration of the universality of the statistical laws of human language has proved a fruitful starting point for identification and investigation of these fundamental principles (4,5). The power of such an approach is illustrated by recent studies of the generality of Zipf's law of abbreviation (6).…”

mentioning

confidence: 99%

See 1 more Smart Citation

Gelada vocal sequences follow Menzerath’s linguistic law

Gustison

Semple

Proc. Natl. Acad. Sci. U.S.A.

et al. 2016

Self Cite

140

Identifying universal principles underpinning diverse natural systems is a key goal of the life sciences. A powerful approach in addressing this goal has been to test whether patterns consistent with linguistic laws are found in nonhuman animals. Menzerath's law is a linguistic law that states that, the larger the construct, the smaller the size of its constituents. Here, to our knowledge, we present the first evidence that Menzerath's law holds in the vocal communication of a nonhuman species. We show that, in vocal sequences of wild male geladas (Theropithecus gelada), construct size (sequence size in number of calls) is negatively correlated with constituent size (duration of calls). Call duration does not vary significantly with position in the sequence, but call sequence composition does change with sequence size and most call types are abbreviated in larger sequences. We also find that intercall intervals follow the same relationship with sequence size as do calls. Finally, we provide formal mathematical support for the idea that Menzerath's law reflects compression-the principle of minimizing the expected length of a code. Our findings suggest that a common principle underpins human and gelada vocal communication, highlighting the value of exploring the applicability of linguistic laws in vocal systems outside the realm of language.communication | language | primate | Menzerath | Zipf

“…Therefore, pressure to reduce dependency lengths implies pressure for compression [25,26], linking a principle of word order with a principle that operates (nonexclusively) on individual words. An understanding of how the principle of dependency length minimization interacts with other highly predictive principles beyond word order is a fundamental component of a general theory of animal behavior that has human language as a particular case.…”

mentioning

confidence: 99%

Towards a theory of word order

Physics of Life Reviews

2017

Liu et al's reflections on the term dependency length minimization [1] may look anecdotal but they are not. By the turn of the 20th century, we put forward a "Euclidean distance minimization" hypothesis for the distance between syntactically linked words and various word order phenomena [2,3] 1 . Later on, pressure from language researchers forced us to replace it with terms such as "online memory minimization" [5] because our initial formulation was obscure to them. Recently, researchers from all over the world have been granted to use the term "dependency length minimization" by the popes thanks to whom [6] came into light. Although "length" is a particular case of distance in this context and thus downsizes our original formulation, it is still abstract enough to allow for progress in theoretical research [7] and frees us from the heavy burden of contingency, i.e. the real implementation of the principle (at present believed to result from decay and interference as reviewed by Liu et al) or the current view of the architecture of memory [8,9]. Our position is grounded on the high predictive power of that principle per se [5].However, the lower generality of the term "dependency length" can be anEmail address: rferrericancho@cs.upc.edu (R. Ferrer-i-Cancho) 1 These were pieces of our PhD thesis [4] that were submitted for publication before its defense. Preprint submitted to ElsevierJune 16, 2017 obstacle to the construction of a fully-fledged scientific field [10]. First, distance minimization allows one to unify pressure to reduce dependency lengths (still distances) with constraints on word order variation and change arising from a principle of swap distance minimization [11]. "Distance minimization" has therefore a higher predictive power and greater utility in a general theory of communication. Second, distance provides a "formal background" or a "specific background" (following Bunge's terminology [10]) from physics or mathematics such as the theory of geographical or spatial networks (where the syntactic dependency structures of sentences are particular cases in one dimension) [12,13] or the theory for the distance between like elements in sequences (where the couple of words involved in a syntactic dependency are particular cases of like elements) [14]. Therefore we agree with [1] on the convenience of the term distance.A less flashy contribution of [6] has been promoting the need of controlling for sentence length (as a predictor of dependency length in their mixed-effects regression model) in research on dependency length minimization, an important methodological issue [15] that was addressed early [2] but neglected in subsequent research (e.g., [16,17,18]).Liu et al focus their review on the fundamental principle of dependency length minimization but understanding how it interacts with other principles is vital. In 2009, we put forward another fundamental word order principle, i.e. predictability maximization, and presented a theoretical framework culminating in a conflict between dependency lengt...