What benefits does using "baby talk" provide, and how do you maximize those benefits?

Question

When I was doing my initial language development research about 7 years ago, I vaguely remember seeing several (probably non-scientific or blog sources) recommend against using "baby talk", arguing that modeling proper pronunciation would help teach by keeping a consistent end goal for the child's language development.

However, this answer indicates that opinion is divided, and this comment on the same answer indicates that there is research supporting the value of baby talk with babies.

So what is the benefit? How does baby talk help? At what point can/should baby talk be left to move on to "regular talk" full time?

Please provide links to studies or research papers documenting how the benefit was quantified, and the methodologies used, if at all possible.

Answer 1

While there have been studies comparing infant preferences between "baby talk" and "regular talk", finding that "baby talk" was preferred, I'd probably lean more towards the results presented in this study which has controlled for positive affect in speech and found that the preferences followed whichever speech had more relatively positive affect. They've aso observed that speaking in higher pitch wasn't necessary or sufficient to convey this positive affect.

The baby prefers the affectionate voice of the caretaker - now how about the language development? I haven't been able to find much specific research into that, but it seems to be okay both to over-pronounce or elongate vowels presumably because it aids in distinction, and speak in full, longer sentences³, even gramatically complex ones, to a baby because it helps with developing general language skills. (Note that the ability to distinguish elongated vowel sounds as carriers of information starts developing between 8 and 14 months of age.⁴) As for muddling words, I haven't been able to find any data. I'd presume it harmless, as long as the child is overall exposed to varied speech, not solely modified "baby" versions of words.

One thing most studies agree on: babies need someone that talks to them, the more the better (vocabulary and general language comprehension gets noticeably better, this seems to be carried over to later reading scores and literacy⁶). This talking has to be directed towards the child and the rich, varied language provides abundant context - for the child context is pretty much the main way of acquiring vocabulary and learning language in general.

³ Marchman, V. A. and Fernald, A. (2008), Speed of word recognition and vocabulary knowledge in infancy predict cognitive and language outcomes in later childhood. Developmental Science, 11: F9–F16.

⁴ Eon-Suk Ko and Melanie Soderstrom and James Morgan: Development of perceptual sensitivity to extrinsic vowel duration in infants learning American English. The Journal of the Acoustical Society of America. Published Online: September 2009

⁶ Fernald, A. and Hurtado, N. (2006), Names in frames: infants interpret words in sentence frames faster than words in isolation. Developmental Science, 9: F33–F40.

Answer 2

As @user25972 pointed out, it's not necessarily "baby talk" that babies prefer, but in one study (cited in the answer) it was positive emotion rather than the higher pitch of baby talk (or infant directed speech/IDS) that matters. However, there is a wealth of information on the preference of babies for baby talk over adult-directed talk (normal speech):

The most robust finding on infants' listening preferences has been widely characterized as a preference for baby talk (BT) over adult-directed speech (ADS). ¹

The main point of baby talk, as offensive as it might be to adults, is that it is generally regarded as happy or positive. In fact, one study showed that the "sad" baby talk of depressed mothers may have a lasting impact on their children²:

‘Baby-talk’ is common across cultures. It underpins infant vocal preferences, and helps regulate infant engagement. Its longer-term significance is unclear. In a longitudinal study, we found indications of ‘sadness’ in postnatally depressed mothers’ baby-talk statistically mediated effects of maternal depression on offspring adolescent affective disorder. (emphasis mine)

Think about baby talk: facial expressions are usually positive and exaggerated. Pitch varies more than in ADS. It is more interesting to the baby to look at (more variation in visual cues) and it teaches something about social interaction. Vocabulary is more limited,sentence structure is shorter and more repetitive, allowing easier learning of phonemes, etc.³:

Studies now suggest that infant-directed speech (a) promotes infant attention to language, (b) fosters social interaction between infants and caregivers, and (c) informs infants about various aspects of their native language by heightening distinctions relative to the speech addressed to adults. New developments focusing on the social role of infant-directed conversational interactions highlight the importance of caregiver responsiveness to the infant. Building a communicative foundation even prior to the time language emerges is crucial for fostering language development.

Furthermore, IDS is exaggerated, meaning vowel sounds are easier to pick out:

One of the most widely cited characteristics of IDS speech is the expanded “vowel triangle". When plotted in perceptual space, vowels in IDS are farther apart than the same vowels in ADS. This larger acoustic difference has been argued to simplify infants’ task of constructing vowel categories (Kuhl et al., 1997; although cf. McMurray, Kovack-Lesh, Goodwin, & McEchron, 2013).

Social aspects of baby talk should be self-evident. When a baby says, "aahgooo" to a parent, the usual response is a delighted "Aahgooo" (IDS) back, not, "What are you trying to tell me, Jennifer? Are you hungry? What makes you say aahgooo?" The former is social interaction the baby can appreciate; the latter is confusing gobbledygook to the baby.

These attentional findings are supported by research using neurological and brain imaging methods showing that IDS results in more brain activation than ADS—for example, in infants’ left and right temporal areas (Naoi et al., 2012) and frontal lobes (Saito et al., 2007). IDS elicits increased neural activity (i.e., larger event-related potential responses) from both 6- and 13-month-olds between 600 and 800 milliseconds (N600–800), which is related to attentional processing (Zangl & Mills, 2007).

I'm limited to a phone right now, and as I am a dinosaur, constructing this answer is not easy. I will end it here, hoping that the case for baby talk has been made.

_{1 Infants' Listening Preferences: Baby Talk or Happy Talk?
2 Sadness in mothers’ ‘baby-talk’ predicts affective disorder in adolescent offspring
3 (Baby)Talk to Me: The Social Context of Infant-Directed Speech and Its Effects on Early Language Acquisition, Roberta Michnick Golinkoff, Dilara Deniz Can, Melanie Soderstrom, and Kathy Hirsh-Pasek, School of Education, University of Delaware; Institute for Learning and Brain Sciences, University of Washington; Department of Psychology, University of Manitoba; and Department of Psychology, Temple University, PDF}