The cognitive science of infant language: What’s a parent to do?

This was the most fun paper of my master's degree.  I combined the advice from many parenting books with evidence from research literature, in hopes of giving parents a solid grasp of how influential our parenting strategies are (or not) and how to change our behavior to more fully stimulate a baby cognitively.

The short of it is this: babies can't learn language from televisions or toys that talk, but if the parent repeats the television or toy, babies can learn it.  Babies learn far faster when parents talk about what the baby is already looking at instead of trying to distract them.  Most importantly, there's no need to feel pressured to do anything differently whatsoever -- your brain and instinctive behavior match your baby's needs naturally in a breathtakingly orchestrated way.

I hope it's as fun to read as it was to write.  Check out the books in the reference section if you want to know more about the neurology of infant cognition.

******


1.0 Introduction

“It’s so annoying when parents try to get their babies to speak bunches of languages and stuff. If your kid’s going to be smart, they’re going to be smart, and if they’re not, they’re not. There’s nothing you can do.” Thus opined a fellow mother upon hearing me speak to my 4-month-old in Spanish. It got me thinking about the influence parenting strategies might, or might not, have on an infant’s linguistic and cognitive ability, and I began to question the motivation behind some of my own behaviors with my infant. The American culture holds linguistic and cognitive achievement as a very high priority, so much so that we tend to believe talking to a preverbal infant constantly is a universal human trait. While it seems absolutely instinctive to us to talk to babies constantly despite the fact that they can’t respond, this is actually a very Western convention, and is no more instinctive than wearing a wedding ring on the left hand (Small, 1998). The behavior reflects our cultural conviction that an individual’s linguistic and cognitive ability can be a key factor in a child’s success in life. This priority is so pervasive that the question “How can I accelerate my child’s cognitive development?” was dubbed “the American Question” by the father of developmental psychology, Jean Piaget (Bringuier, 1980). After thinking about her comment, I decided to soak up such wisdom as is offered by parenting books regarding language, and tie that in to what I have learned from research literature in cognitive psychology. What parental behaviors might such evidence recommend, and, as the fellow mother would question, is there any point attempting to adopt them?

There is no question that a child’s ability to soak up language is breathtakingly complex and blindingly powerful. Researchers have used innovative and insightful nonverbal strategies to test infant language acquisition from many perspectives, including the fields of ethnopediatrics, psychology, neurology and various branches of linguistics. The gestalt of the literature paints a picture of typical babies as brilliantly – and effortlessly – engaged in the feat of mastering systems of spoken and gestural communication.  Surprisingly, unconscious parental behavior often unfolds in neatly orchestrated lockstep with infant needs, as adults subtly modify speech and interactional strategy to scaffold learning. Parental involvement plays a pivotal role in shaping a seedling verbal ability into the grand and convoluted tree it will become, so research literature is replete with implications regarding how parents can interact with a child to maximize that child’s verbal and cognitive skills. This paper provides an overview of ways this research is conducted, what it reveals, what it implies about parenting strategies and how several evidence-based parenting books recommend that knowledge be applied in daily life.

2.0 Methodologies

Before children can speak, they reveal their understanding of their world in different ways. One way is to directly measure brain activity time-locked to the presentation of stimuli, as Molfese et al (1999) did in measuring the linguistic processing of newborns. A more common, less technological method is to measure how an infant perceives and processes her environment is by quantifying habituation to stimuli. An infant is said to habituate to a cue when her interest in it declines, which is measured in different ways for different ages. Some techniques allow a baby to select or maintain a preferred stimulus instead of measuring habituation. Here are a sample of such research methodologies.

2.1 From birth to about 6 months old, babies’ interest in their environment is reflected in their rate of sucking. Pacifiers modified to be data collection devices are used to measure when a baby grows bored of a stimulus or becomes interested in something new. This boundary reflects the categories in a baby’s mind, because something being new requires that it be markedly different from what came before.

2.2 Sucking can also be used to determine how motivated an infant is to maintain an enjoyable stimulus. The data-collection pacifier can “remote control” a video display such that the infant can select which video to see by sucking faster or more slowly. If a baby sucks less than 30 times a minute, the baby might see a woman speaking with a soundtrack that does not match her articulations, if the baby sucks harder than that threshold then the sound track could sync up with the video.

2.3 The strength of the sucking response to interesting stimuli weakens after about 6 months, but closely related methodology called head-turn conditioning can still quantify habituation, as used by Kuhl et al in studies of phonetic perception (Kuhl et al, 2003). The baby will sit on a parent’s lap while looking at an interesting puppet or toy that is kept moving during the repeated presentation of the same stimulus. The infant is trained to turn and look in a different direction every time the stimulus changes in order to see a very attractive toy dance with flashing lights. After being thoroughly trained so she expects the attractive toy to appear only when the cue changes, the baby will continue to watch the moving puppet until she perceives a new category boundary, at which point she turns her head to look for the attractive toy.

2.4 The pacifier-as-remote-control scenario has a parallel at this stage too; a baby can actively change a sound recording (rather than responding to a change in it) by turning her head. One study showed that babies will turn their heads to hear a woman speaking motherese in any language rather than adult-oriented speech in the baby’s primary language (Gopnik et al, 1999).

2.5 Puppets are helpful when investigating the cognitive processes of toddlers who can verbalize their thoughts but only vaguely. These are useful in experiments where there is a dissociation between what a toddler knows implicitly and what a toddler explicitly says, as in Low’s investigations on false-belief understanding (2010). A toddler watching puppets interact will look at certain things depending on her understanding of the situations being represented. This reveals what the toddler thinks different “people” understand when one person has witnessed an event the other puppet did not “see”.

3.0 Findings: Linguistic and cognitive development from birth to parenting

It’s hard to know where to begin in summing up research conclusions about infant language acquisition, but we’ll start with the physiological neurolinguistic development of newborns and proceed through to evidence regarding the impact of caregiver interaction.

What do we bring onto the stage of life? At birth we have more brain cells than at any other time of life; by some estimates nearly half (80 of 200 billion) will have died by the age of 1 (Sunderland, 2006). This is by no means the tragedy it sounds, because the neural proliferation of birth is to cognition what the broad-spectrum static of white noise is to music. The first year of life provides a bonsai-like pruning of brain cells, allowing the static to resolve into symphony by removing, not adding, frequencies of activation. Whether connections abound at this time seems in dispute, some authors claiming the proliferation of newborn neural matter is very unconnected, with individual neurons blindly firing without triggering connections (Sunderland, 2006) while others claim the exact opposite extreme, that the associations between brain cells are far, far more redundant than at any other point in life (Gopnik, 1999). The implications for a newborn’s subjective experience of consciousness are striking, since neural activation absent connections would likely be experienced as senseless random perceptions, but heavily redundant interconnectivity between brain cells would cause a near-complete synesthesia of sensations that are experienced distinctly by adults. Recent research backs up the hyperconnectivity hypothesis, finding that shapes influence color preferences in 2 and 3 month olds but not in 8-monthold infants or adults (Wagner and Dobkins, 2011). Arbitrary linking of perceptual modalities decreases as infants grow older, refining the perceptual conflagration into the more analyzable data stream adults navigate as consciousness.

The linguistic facet of such cognitive sculpting comes on stage immediately, even before time allows experiential diminution of associations. Amazingly, the differences in how a newborn’s brain processes sounds are significantly predictive of reading difficulties at the age of 7. Studies analyzing brainwaves of newborns via event-related potentials have found that the more finely voices are processed (through brainwaves reflective of linguistic activation), the less likely the baby is to have reading difficulties years later (Molfese et al, 1999). The neurological equipment with which we come to the game has a strong impact on the ease with which we will develop linguistic ability, but the richness of our environment can trump our original aptitude – in a good way or a bad one.

It does this via the bonsai-like neural pruning during the first year. This is led by a re-weighting of connections based on how often environmental stimuli link brain activity in different areas. A major finding that overturned expectations of experts in the field was that of infant phonetic discrimination, whose development directly reflects the frequency of an infant’s exposure to speech sounds.  Researchers long believed that babies had to learn to hear distinctions in sounds, that hearing reflected vision in beginning fuzzy and only becoming distinct with exposure to language (Gopnik, 1999). In fact babies begin as “citizens of the world” (Kuhl,2003), capable of differentiating sounds in all languages the way dogs can. Toward the end of the first year, between 6 and 12 months, a specialization occurs, and the ability to discriminate foreign-language sounds sharply declines. Instead of hearing every articulatory variation as a distinct sound, babies begin to perceive a spectrum of sounds categorically. Sounds that trigger contrastive meaning (phonemes) in the language(s) of a baby’s environment receive special status as prototypes, until phoneme boundaries modify perception categorically. A frequency heuristic seems to be at work, building up prototypes and having similar stimuli trigger categorical recognition.

Patricia Kuhl found that if a baby is consistently exposed to a foreign language at this point, whatever she has heard before, then her brain will maintain the distinctions of the sounds in that language. She exposed 9 month old infants to 12 sessions of foreign languages, each 25 minutes long, over a one month period. Amazingly, such brief exposure served to preserve discriminatory categorical perception in non-native speech sounds, but only if the exposure occurs with a live tutor. Watching people talk on television or listening to CDs has no effect (Kuhl, 2003).

This differential influence of the source of language shows up in other studies as well. A sound has to come from a mouth instead of any sort of technological device in order for an infant to consider it a candidate for a label. Both words and animal noises were subject to the linguistic process of mutual exclusivity in the behavior of 20-month-olds, but only when they were spoken by a researcher in the room rather than a hand-held noisemaker (Brojde and Colunga).

Together these imply two things, to my thinking: there’s some linguistic “social brain” switch that has to be turned on in order for an infant to process sound as language. This leads to the conclusion that 2) all of the stuffed animals that say “Hola! Como estas?” and all the Baby Einstein-genre videos purporting to teach infants foreign languages are actually not being processed as language by babies at all. An implication for parental behavior becomes obvious here: if parents were to repeat the words the adult hears from the toy or video to which they are exposing their little one, the baby would then process it as language. The electric voice won’t be heard as a voice by the infant, but if the parent echoes it in person, the baby will understand it not as environmentally-generated noise but as meaningful socially relevant communication. I apply this by trying to sign along with Singing Time videos designed for children when my 10-month-old Amelie and I watch them. Very often she focuses on the movement of my hands with the same minute-seeming precision she uses on occasion to watch my lips as I speak.

Sign language is an interesting area of dispute in regards to what effects it has on infant linguistic development. The formal research literature has generally found it to be not at all as empirically beneficial as parents are told it is (Keister and Keister, 2009). For the most part such claims are commercially oriented, not backed by solid research, and thought by many to be manipulative, encouraging parents to pay money for programs that actually just formalize the gestural communication all preverbal infants establish with their parents. It is really more a nominative label-oriented gestural code, lacking the rich special syntax unique to manual languages (Sacks, 1989).

While the benefits of teaching an infant codified sign language have not been scientifically verified as such, an infant’s ability to communicate does have the benefit of lowering frustration by being able to signal the cause of discontent (Keister and Keister, 2009). The primary benefit however is the environmental focus on rich interactive connection. The sort of parental interaction that goes along with familial efforts to establish early communication is itself likely to benefit an infant’s cognitive development. When parents engage in intensive sustained collaborative focus with an infant, it seems to throw the social switch that tells a baby’s brain that environmental information is important. When an adult pays attention for a long time to the same thing the baby is paying attention to and elaborates verbally on the object of the baby’s focus, the baby’s vocabulary acquisition is accelerated. A study of deaf and hearing children with deaf and hearing mothers showed that when mothers comment on an object the child is already paying attention to, when they elaborate on jointly sustained attention, the child is more likely to learn words. When we pay attention to something with a baby, the baby’s brain gets ready to accept a sound or a symbol as a label for the object. To me it makes sense that beginning the signifier/signified connection earlier would open the door further for later vocabulary acquisition, and understanding the pragmatics of turn-taking would help an infant develop interpersonal sensitivity.

Whether using a visual or spoken communicative system, the language in which we encode an event impacts the details we are most likely to remember. In Spanish, the agent of an unintentional event (dropping) is not linguistically coded and can be forgotten (Fausey & Boroditsky, 2010). Language acts like a container into which we drop an idea, and the form of the container shapes our recall of the idea. Oliver Sacks in Seeing Voices reports on data from interviews with adults who, due to congenital deafness and late acquisition of language, and he claims that the ability to predicate linguistically is so essential to our cognition that with out language we exist in a continuous river of timeless now. Such adults often remember a stream of identity, with no ability to remember that yesterday the sun was shining, no way to form abstractions and learn. Studies with adults show that the ability to make finer distinctions (as in having been taught more detailed bird-species classification) allows larger distinctions to be made more easily, possibly because the more adept we become at thinking of a descriptor as independent of the described (being able to think of “red” instead of “the red car”) the more powerfully we can manipulate abstraction. If Sacks’ conclusions are accurate, language necessarily underscores the logical and analytical ability that makes us most human, because in its absence we cannot hold information long enough to process it or deduce implications. This is absolutely fascinating because it indicates directly that without other humans, we fail to become truly human ourselves; interpersonal interaction is literally a mandatory part of realizing the genetic aptitude afforded by the singular evolution of our species. In a very real sense we are a single organism, necessary to each other in order to manifest the complete richness of our abilities. Interpersonal interaction is encoded in our genes, and we require a community to become fully ourselves.

This community finds its first embodiment as family and actually has a very directive impact on the linguistic and cognitive sculpting of a child’s mind. The good news is that families do this almost entirely unconsciously. Caregivers modify linguistic forms in many ways to make learning the language easier. We exaggerate the differences in phonemes, making the prototype/histogram project easier; we use engaging and sweeping changes of prosody so the child can pay attention longer (Gopnik, 1999). Caregivers also modify the complexity of sentences in order to help a child first master a new word, using it in more and more simple situations right before a baby gets it and then making the language complex again (Roy, 2011). Tight feedback systems must dictate caregiver speech habits unconsciously to cater to the needs of the burgeoning speaker. So much of this adult language modification is instinctively unintentional that Gopnik (1999) concludes we have no need of expert advice, only time to do what we naturally do as we interact with children.

Such untrained interaction has been found to markedly influence cognitive development in infants, so much so that maternal behavior with a 6-month-old has been found in several studies to predict performance on intelligence assessments given when the same child is 4 years old (Bornstein, 1985). Didactic parental activities are extremely influential in promoting (or, it may follow, inhibiting) the linguistic and cognitive achievement of their children. The maternal habits that proved most  efficacious in producing advanced linguistic ability were founded (as previously mentioned studies would imply) on intensive sustained attention. The most powerful way to help an infant develop advanced linguistic skills is to frequently encourage them to pay attention to properties, objects and events in the home environment. Such attentional allocation is shared between infant and parent, awakening the social processing of the brain, expanding the labels available for manipulation and by so doing honing an infant’s ability to abstract a trait from the object it describes. In such a way we can empower our children to think clearly and specifically, to use language as a multi-faceted tool in analyzing the environment and making appropriate decisions based on available information.

4.0 Conclusion

Was my fellow mother right in claiming that an individual’s linguistic and cognitive ability are set in stone from the moment of birth? Absolutely not: scientific literature is replete with data contradicting that stance, and there is in fact very much we can do. Every experience from the moment of birth influences the trajectory of neural pruning, and the early developmental paths parents set for their infants may heavily influence what paths are available to that child later in life (Gopnik, 1999).

However, that frustrated mother was reacting understandably to inappropriate societal pressure regarding overextension for the linguistic and cognitive benefit of a baby. Her implication that there is no fundamental need to go out of our way to produce a rich linguistic environment for a child is true, because we as parents have co-evolved with babies, and a standard loving parent’s unconscious interactional strategies will equip a healthy infant with strong linguistic and cognitive ability. All the technology marketed to parents as enhancing an infant’s wellbeing does so without the slightest shred of supporting evidence that the claims are true. Part of the usefulness of science is that it inoculates us against pseudo-science, and learning may best be accelerated by encouraging an involved, attentive, happy baby to interact with the people she loves (Gopnik, 1999).

What has all of this information had on my choices as a parent and my understanding of my own  behavior? I am now very relaxed and confident that my instinctive linguistic style is perfect for my child. Instead of systematically creating and repeating situations that would allow her to link the same sound with the same object as a strategy of encouraging early word recognition, I speak to her nearly every day in Spanish, French, English, Arapaho and American Sign Language. I do it not because I am vested in her development of extreme linguistic ability or cognitive abstraction; I find it likely that such inconsistent verbal data will throw off her histogram of phoneme frequency, and my utter lack of consistent routine may well temporarily impede her mapping of signifier to signified. The fact that we listen to recordings of languages instead of music in my house is no doubt changing in some way how her brain will process music, perhaps indicating that an intentional focus on musical training at a later age will be wise.

After all the reflection on the impact my parenting will have on her cognition, my ultimate conclusion is this: so what? There is no reason whatsoever to be worried or feel pressured about the impact my behavior will have on Amelie’s cognitive and linguistic abilities, and I am not motivated (at least not consciously) by any desire to produce a certain outcome, which means I will not be disappointed if her verbal ability proves sub-par. I do all of this because it’s fun, every word of it, beginning with the studying of Italian for hours every day during the last trimester of pregnancy and continuing through our most recent joint endeavor of home-schooling a 3-year-old in French one day a week. It’s all a hobby, all an experiment whose results can only be interesting, only delightful, only thought-provoking. And in any case, last week at the age of 10 months she stopped crying in her car seat mid-wail when she heard the Italian lessons I repeated nine zillion times in the third trimester of pregnancy. Whatever impact my behavior has on her future, it’s wonderfully fulfilling to play the linguistic game to its fullest now.

5.0 Bibliography

Brojde and Colunga. Out of the Mouths of Babes: The effect of source on 20-month-olds' Use of Mutual Exclusivity. Department of Psychology, University of Colorado.

Bornstein, M. H. "How Infant and Mother Jointly Contribute to Developing Cognitive Competence in the Child." Proceedings of the National Academy of Sciences 82.21 (1985): 7470-473.

Bringuier, Jean-Claude. "Conversations with Jean Piaget." Society 17.3 (1980): 56-61.

Fausey, Caitlin M., and Lera Boroditsky. "Who Dunnit? Cross-linguistic Differences in Eye-witness Memory." Psychonomic Bulletin & Review 18.1 (2011): 150-57.

Gopnik, Alison, Andrew N. Meltzoff, and Patricia K. Kuhl. The Scientist in the Crib: Minds, Brains, and How Children Learn. New York: William Morrow &, 1999.

Kiester, Sally, and Edwin Kiester. The Secret Language of Babies: The Body Language of Little  Bodies. London: New Burlington, 2009.

Kuhl, P. K. "Foreign-language Experience in Infancy: Effects of Short-term Exposure and Social Interaction on Phonetic Learning." Proceedings of the National Academy of Sciences 100.15 (2003): 9096-101.

Low, Jason. "Preschoolers’ Implicit and Explicit False-Belief Understanding: Relations With Complex Syntactical Mastery." Child Development 81.2 (2010): 597-615.

Molfese, Dennis L., Victoria J. Molfese, and Kimberly A. Espy. "The Predictive Use of Event-Related Potentials in Language Development and the Treatment of Language Disorders." Developmental Neuropsychology 16.3 (1999): 373-77.

Roy, Deb (2011). TED talk: The Birth of a Word.
http://www.ted.com/talk/deb_roy_the_birth_of_a_word.html

Sacks, Oliver W. Seeing Voices: A Journey into the World of the Deaf. Berkeley: University of California, 1989.

Small, Meredith F. Our Babies, Ourselves: How Biology and Culture Shape the Way We Parent. New York: Anchor, 1998.

Sunderland, Margot. The Science of Parenting: How Today's Brain Research Can Help You Raise Happy, Emotionally Balanced Children. New York, NY: Dorling Kindersley, 2006.

Wagner, K., and K. R. Dobkins. "Synaesthetic Associations Decrease During Infancy." Psychological Science 22.8 (2011): 1067-072.