Skip to main content
NCBI home page
PLOS One logoLink to PLOS One
. 2021 Jan 7;16(1):e0244968. doi: 10.1371/journal.pone.0244968

Semantic priming supports infants’ ability to learn names of unseen objects

Elena Luchkina 1,2,*,#, Sandra R Waxman 1,2,#
Editor: Lucas Payne Butler3
PMCID: PMC7790528  PMID: 33412565

Abstract

Human language permits us to call to mind objects, events, and ideas that we cannot witness directly. This capacity rests upon abstract verbal reference: the appreciation that words are linked to mental representations that can be established, retrieved and modified, even when the entities to which a word refers is perceptually unavailable. Although establishing verbal reference is a pivotal achievement, questions concerning its developmental origins remain. To address this gap, we investigate infants’ ability to establish a representation of an object, hidden from view, from language input alone. In two experiments, 15-month-olds (N = 72) and 12-month-olds (N = 72) watch as an actor names three familiar, visible objects; she then provides a novel name for a fourth, hidden fully from infants’ view. In the Semantic Priming condition, the visible familiar objects all belong to the same semantic neighborhood (e.g., apple, banana, orange). In the No Priming condition, the objects are drawn from different semantic neighborhoods (e.g., apple, shoe, car). At test infants view two objects. If infants can use the naming information alone to identify the likely referent, then infants in the Semantic Priming, but not in the No Priming condition, will successfully infer the referent of the fourth (hidden) object. Brief summary of results here. Implications for the development of abstract verbal reference will be discussed.

Introduction

Language is among our most powerful tools for learning and communication. It permits us to learn information that does not, or cannot, manifest perceptually at the time of learning [1], such as historic facts, hypothetical scenarios, or scientific constructs. From toddlers hearing words for absent and often unknown objects (e.g., “Daddy outside is fixing the trellis”) to physicists conversing about gravity, the communicative power of language enables us to transmit new information across space and time. This capacity rests upon an appreciation that language can convey information that is decoupled from the here-and-now. This uniquely human power is enabled by abstract verbal reference–an appreciation that words are linked to mental representations that can be established, retrieved and modified, even when the entities to which they refer are not perceptually available [2–4]. How, and how early, do infants gain this appreciation?

Evidence documenting infants’ ability to comprehend displaced (or fragmented) reference (an operational definition of abstract verbal reference, which entails reference to objects or events that were recently visible but subsequently removed from infants’ view) provides valuable insight into these questions. A nascent appreciation of displaced, or fragmented, reference appears to be present by 12 months of age. Infants interpret pointing gestures as referential [5,6] and refer to a recently hidden entity by pointing to its previous location [7–12]. In addition, 12-month-olds understand that words can refer to recently hidden objects [13–15]. By 14 months, infants bring greater social knowledge to bear, responding to a speaker’s verbal request for an absent object by selecting an object that a speaker prefers [16]. Infants at 14 months also use words to retrieve memories of past events [17]. They remember properties of named, but absent, objects [18] and produce communicative behaviors towards the location of recently hidden objects [19].

In each of these preceding instances, 12-to-14-month-olds’ resolution to the puzzle of displaced reference hinges upon their use of recent visual “anchors”–reminders that scaffold their representation of the now-absent referent [20]. By 16 months, infants go further, successfully resolving displaced reference on the basis of verbal testimony alone [21], even if the absent entity has not previously been “anchored” within the context of the experiment [22]. However, in each of these studies, infants can rely on existing knowledge, acquired when those entities were visible.

What remains unknown is when infants successfully acquire new representations and update them on the basis of language alone. This is the signature of abstract verbal reference. We propose that if infants can establish a mental representation for the meaning of a novel word from language alone and later draw on that representation to identify a candidate referent of that novel word, this will constitute clear evidence for abstract verbal reference.

There is considerable evidence that 19- to 24-month-old infants resolve abstract verbal reference and successfully infer the meanings of novel words on the basis of linguistic information alone, in the absence of any visual referents [23–26]. For example, Ferguson et al. [2014; 27], showed that 19-month-old infants successfully identify the referent of a novel word even if that referent is not available when the name is introduced. In their study, 15- and 19-month-olds were presented with novel nouns that were arguments of known verbs, either animacy-selecting (“The dax is crying”) or animacy-neutral predicates (“The dax is right there”). When the novel word was presented in linguistic constructions that specified animate arguments, 19-month-olds (but not 15-month-olds) successfully mapped the novel word to an animate object when it later became visible.

But why did 15-month-olds fail? The reviewed literature suggests that between 14 and 16 months, infants can resolve reference to absent objects. On possibility is they cannot yet establish new mental representations for novel words based on linguistic information alone. However, it is also possible that they do have the requisite linguistic and conceptual capacities, but that their relatively sparse lexical knowledge masks them (on average, infants in Ferguson et al. comprehended 4.3 verbs from the MacArthur Level II Short Form). For example, in Ferguson’s task, infants were required to infer the meaning of a novel noun based on a known verb. Infants’ verb knowledge at 15 months may have been too sparse to support the acquisition of novel noun arguments.

We therefore introduce a new paradigm that taps into infants’ capacity for abstract reference, but relies instead on the nouns they know [on average, 15-month-olds comprehend 71 nouns; 28] and their emerging sensitivity to semantic neighborhoods [29,30]. We ask whether priming a sematic neighborhood supports infants’ ability to establish a representation for the referent of a novel noun, even when the referent cannot be seen. At issue is [1] how early infants successfully resolve abstract verbal reference, as measured by their ability to acquire novel word meanings from language alone, and [2] whether semantic priming is supports such learning.

Experiment 1

The goal is to assess whether 15- to 16-month-olds can successfully establish a representation of a novel word’s meaning from language input alone, and recruit this representation later to identify a referent of the novel word. We focused on infants aged 15–16 months because although there is ample evidence that they successfully resolve displaced reference for known words [16,18,22,27,31], there is no evidence that they learn novel word meanings without a referent being present.

Method

The proposed work conforms with the US Federal Policy for the Protection of Human Subjects and Northwestern University Human Research Protection Program requirements. All research personnel have completed the required responsible conduct of research training and has been certified by the Collaborative Institutional Training Initiative. A protocol amendment that extends the scope of the protocol to include the proposed procedure was submitted and approved by the Institutional Review Board of Northwestern University. This research will be conducted based on caregivers’ written consent and participants’ willingness to complete the experiments. All participants will receive compensation, regardless of successfully completing the experiments.

Participants

Seventy-two (N = 24 per condition) infants from the Chicago area will be recruited from the laboratory database. Infants will have no more than 15% exposure languages other than English. Caregivers will complete a MacArthur Short Form Vocabulary Checklist: Level II [Form A; 32], augmented with words used in the experiments but not included in the form (see S1 Appendix). They will also complete a demographic questionnaire about their education, employment, gender, and race (see S2 Appendix).

The number of participants was determined based on power analysis performed with G*Power 3.1 [33] for a repeated measures between-subject F-test with target effect size f2 = .1225 (medium; [34], alpha = .05, power = .8, number of groups (experimental conditions) = 3, number of trials = 4, and projected correlation between measurements = .5. We reasoned that compelling evidence would require at least a medium effect size of experimental condition. Based on these parameters, the required number of participants is 72. The planned analyses include Generalized Linear Mixed Model rather than an F-test. We used F-test-based effect size estimations for simplicity. The four test trials were treated as repeated measures.

Participants may be excluded and replaced due to fussiness, crying, or equipment failure. Further, because our Covid-19 Contingency Plan (see below) involves testing infants using Lookit–an online platform designed for developmental research–we expect a data loss rate of about 47–50% [35] due to infants’ crying or fussiness, parental interference, or technological issues. This, we anticipate that the total number of infants recruited could reach 140.

Timeline

We intend to begin data collection as soon as the current registered report is granted in-principle acceptance and project that all data will be collected and analyzed within a calendar year from that date. If the results of the Stage 1 review warrant major revision of the proposed experimental procedures, we intend to resubmit the report for Stage 1 review within one month after receiving reviewers’ feedback.

COVID-19 contingency plan

The global COVID-19 pandemic is currently preventing us from testing participants in the lab. We are transitioning the experiments in this report to Lookit–an online platform for developmental research. Using this platform, we will be able to record participants’ eye movements using web-cameras in their desktop or laptop computers and subsequently code their looking behavior manually. While the reportedly higher data loss rate may slow our data collection, Lookit’s access to a larger sample of participants and at-home participation is likely to offset its associated data loss.

Apparatus

A Tobii T60XL corneal-reflection eyetracker will be used for stimulus presentation and data collection. The eyetracker has a sampling rate of 60 Hz, and a display size of 57.3 × 45 cm.

Stimuli

Infants will view a series of four video-taped vignettes, each featuring a new novel word-object pairing. Each video begins with an actor pointing three familiar objects, naming each with its familiar basic-level noun. Next the actor looks toward an object hidden behind her back, naming it with a novel word. See Fig 1 for a representation of a single trial and S3 Appendix for a complete list of stimuli. All stimuli will be made available on Open Science Framework.

Fig 1. A representative example of visual and linguistic information presented in each phase in the semantic priming condition.

Fig 1

Open in a new tab

Stimulus selection and design

We used vocabulary norms [36] to select the visual and linguistic materials. For the familiar objects presented in the Priming phase, we selected objects whose names are understood by at least 60% of 15-month-olds (nouns: apple, banana, orange, jacket, sock, hat, truck, car, bus, cat, dog, horse). For the novel objects presented during Test, we (a) selected objects whose names infants do not understand (e.g., rare tropical fruits, exotic animals), and (b) chose pairs of objects from distinct semantic neighborhoods (e.g., papaya vs aardvark).

Procedure

All infants will participate in 4 trials, each including a Priming and a Test phase (Fig 1). Each trial will include one novel object-novel word pairing. Infants will be assigned randomly to condition. What varies across conditions is the first three objects presented during Priming: in the Semantic Priming and Follow-up Control conditions, these are all members of the same semantic neighborhood (e.g., animals, food, clothing); in the No Priming condition, they are from different semantic neighborhoods. We selected a between-subject design to avoid potential carry-over effects, which would dilute the predicted advantage of Semantic Priming. Another solution to minimize carry-over effects might be to increase the number of trials and to present them as blocks. However, doing so would lengthen the procedure and therefore runs the risk of exceeding infants’ attention. In all conditions we provided infants with rich communicative and referential cues to support their interpretation of Trial 4 of the Priming phase, in which the object is hidden, as a labeling episode as well.

Semantic Priming condition (N = 24; Fig 1): Priming (35 s): During Priming an actor will point to and name three familiar objects, all from the same semantic neighborhood (fruits, clothing, vehicles, animals). The objects will appear one at a time, behind the actor; the actor will turn to point to and name them, using their familiar basic-level nouns (e.g., “Look! An apple! Do you see the apple?”). A fourth object, seemingly unintentionally occluded by the actor’s body, will be labeled with a novel noun (e.g. “Look, a modi!”), creating a situation of abstract reference—during naming, no referent object is visible. The actor will alternate between looking to the object and to the camera to indicate that the referent object is located behind her. We present infants with familiar word-object pairs during the priming phase to access their resident semantic knowledge [30,37–39].

Test (10 s): During Test, infants first will see an attention getter (2 s), followed by a blank white screen (2 s), during which they will hear the first prompt to look to the object corresponding to the novel label, e.g., “Now look! Where is the modi?”. Immediately after, two objects will appear side-by-side—one is a member of the same semantic neighborhood during the Priming phase (e.g., a novel fruit) and the other is a semantically distant item (e.g., a novel furniture item). With these objects visible, infants will hear, e.g., “Can you find the modi?”.

No Priming condition (N = 24; Fig 2): The procedure will be identical to the Semantic Priming condition, except that during Priming, infants will be presented with familiar word-object pairs, all from different semantic neighborhoods (e.g., a hat, truck, and horse). Thus, no priming of any particular semantic neighborhood will be involved.

Fig 2. A representative example of visual and linguistic information presented in each phase in the no priming condition.

Fig 2

Open in a new tab

Follow-up Control condition (N = 24): Although our pilot data (see below) suggest that 15-month-olds successfully establish novel object representations based on language input alone, it is important to rule out the possibility that their success is driven by the semantic relations among the objects themselves and therefore not dependent upon infants’ inferences about novel word meanings. We therefore plan a Follow-up Control condition, identical to the Semantic Priming condition with one exception: during Test, infants will hear a novel word that differs from that presented on Trial 4. If infants’ object preference at test is driven by the objects themselves, rather than naming, then performance in this Follow-up Control should not differ from that in the Semantic Priming condition. Conversely, if performance in the Semantic Priming condition is a consequence of verbal reference, then performance in this Follow-up Control condition should not differ from that in the No Priming condition.

Data preparation

We will analyze infants’ looking behavior from the onset of the image presentation through the end of the trial (6 s). Our decision to conduct 6 s of image presentation in test trials is based on prior evidence: Ferguson et al. (2014) [27] tested infants of comparable age in a procedure that also involved establishing word meanings in the absence of a visible referent. As in Ferguson et al., trials on which the total looking to the screen is less than 1s will be excluded from analyses.

For data collected in the lab using a Tobii T60XL eyetracker, we identify areas of interest (AOI) of approximately 720*790 pixels around each of the object images, such that every AOI includes a 50-pixel-wide area around an object image. This is done to accommodate for imprecisions of the eyetracking data, which often result from infants’ head movement or blinking. Gazes outside these areas will be excluded from analysis, as will be trials in which an infant’s total looking time is less than 1 out of 6 seconds.

For data collected on Lookit, manual frame-by-frame coding will be performed by trained research assistants who will watch video-recordings of online testing sessions and determine whether the infant is looking to the left or to the right side of the screen during each frame. Depending on the location of the target object, left-right codes will then be converted to target-distractor codes by a different research assistant. Ten percent of all videos will be selected for reliability coding. All research assistants involved in video coding will be agnostic to experimental conditions or counterbalancing schemes.

For each trial of each infant, we will calculate the proportion of looking time devoted to the target [looking to target/(looking to target + distractor)] throughout the analysis window. This should yield up to 4 data points for each infant (one per each of 4 trials). To assess the time course of infants’ attention to the animal and artefact images, eyetracker data will be aggregated by condition into a series of 100-ms bins. For data collected on Lookit, we will use 200-ms bins. Web-cameras typically have sampling rates varying between 30HZ and 60HZ, which translates to frame duration ranging from 16.67 ms to 33.33 ms. Binning our data into 200-ms segments allows us to have at least 6 frames in each bin and to calculate the proportion of infants’ looking to the target image. Fewer frames per bin would result in substantially cruder estimations of infants’ proportion of looking to the target.

Analyses

The dependent variables (DVs) will be [1] infants’ proportion of looking time (LT) to the target image from the image display onset until 6 s later, averaged across four trials; [2] the time course of the proportion of LT for each 100-ms (or 200-ms) bin in the 6-s window. Because our goal is to evaluate the effect of semantic priming on infants’ word learning under the conditions of displaced reference, we will use Condition as the main independent variable (IV).

In analyzing the overall proportion of LT to the target, we will use a Generalized Linear Mixed Model (GLMM) with Condition as a fixed factor and Participant and Test Item as a random factors. We include the Test Item variable to account for the potential effects of infants’ knowledge in different domains–fruits, vehicles, clothing, and animals. We plan to use an identity link, which will require that we test the distribution of the data for normality. We will implement this by running a Shapiro-Wilk test. If the distribution of our data significantly deviates from normal, an appropriate transformation will be applied. In analyzing the time course, we will use the cluster-based permutation analysis [40] to identify significant divergences between the conditions that correspond to mention of the target word.

The effects of demographic factors, vocabulary scores, and trial order will be tested in a preliminary analysis, implemented by fitting a GLMM to the DVs. Trial order is included to test whether infants’ looking preferences change over the duration of the procedure, as they might lose interest or exhibit a learning effect as the experiment progresses. Only the factors that produce significant effects in the preliminary analysis will be included in the main analyses. No other analyses are planned for the data collected via the demographic survey.

Predictions

Because infants as young as 12 months infer the presence of a hidden object by following a speaker’s line of regard [41], understand that words communicate about mental states [42–44], and appreciate referential cues [5], we make the following predictions.

Proportion of LT to the target

Semantic Priming condition: We predict significant effects of Condition, with average proportion of LT (across 6 s) to the target significantly higher in the Semantic Priming than in the No Priming and Follow-up Control conditions. Further, we predict that if infants successfully infer the meaning of the novel word and establish a representation of its referent in the Semantic Priming condition, their LT to the target will be significantly above chance (50%).

No Priming condition: If infants’ looking behavior is influenced by semantic inferences, rather than baseline preferences, then their LT to both objects in the No Priming condition will be equivalent and will not deviate significantly from chance. We predict this because there is no basis for infants to infer the meaning of the novel word after being presented with three semantically distant word-object pairs.

Follow-up Control condition: If infants’ looking preference in the Semantic Priming condition is the result of the semantic relations among the objects themselves, then their performance in the Follow-up Control condition will not differ from the Semantic Priming condition. Conversely, if infants’ looking in the Semantic Priming condition reflects infants’ inferences about novel word meanings, their LT in the Follow-up Control condition should not significantly deviate from chance.

Time course

We expect that the time course data will show significant divergences between the conditions after each verbal prompt. Based on our pilot data, we envision that after each prompt, infants in the Semantic Priming condition will exhibit increased proportion of looking to the target, which we predict to be above the chance level. In the Follow-up Control condition, infants’ looking behavior is likely to not be significantly affected by the verbal prompts because according to our hypothesis, they will have no particular expectations about the meaning of the novel word. It is, however, also possible that they will exhibit a weak preference for the target object (the object that is target in the Semantic Priming condition) based on their expectations about the next item shown given the relatedness of the three demonstrated objects. In the No Priming condition, infants have no basis to form any particular expectation about the object or the meaning of its label. For this reason, we expect that infants’ looking behavior will not be affects by the verbal prompts and remain around the chance level for the duration of the test trial.

Pilot data

The results of the pilot data analyses (N = 18) in the Semantic Priming condition reveal a significant looking preference for the target. Infants’ LT to the target image across 6 s was significantly above chance (50%), M = 64%, SD = 8%, t(17) = 5.5, p < .001. The time course data reflects infants’ reactions to verbal prompts: infants’ looking to the target increased after each mention of the target word (Fig 3). These data provide assurances that the task is well-suited to test 15-month-olds’ ability to learn novel word meanings from displaced reference and provides hints that infants might indeed be using semantic priming to infer such meanings.

Fig 3. Time course of the pilot data (N = 18) in the semantic priming condition.

Fig 3

Open in a new tab

Results and discussion

We reasoned that if 15-month-olds can establish novel word-object mappings from language alone, it would indicate that they can resolve abstract verbal reference. Alternatively, failure on this task will raise several possible interpretations. First, 15-month-old infants may be sensitive to semantic priming [29], but not sufficiently robust to support learning a novel word in the absence of its referent. Second, infants may fail to infer that there is an object behind the actor. If this is the case, further investigation will be necessary to clarify the presence of a hidden object. Third, it is possible that 15-month-olds can resolve abstract verbal reference, but that their exposure to the novel word during priming is insufficient for them to establish novel word-object mappings from displaced reference. These possibilities all warrant follow-up investigations, which will shed light on the development of infants’ inferential and representational capacities and their ability to use such capacities in word learning. We have designed such follow-up studies and will conduct them if 15-month-olds fail to learn the meanings of novel words in Experiment 1. However, based on the pilot data, we predict that these infants will succeed, which leads us to Experiment 2.

To be expanded when data collection and analyses are complete.

Experiment 2

If the results of Experiment 1 suggest that at 15 months, infants successfully establish novel word-object mappings from language alone, then we will go on to identify entry points for this capacity. More specifically, we will focus on 12-month-old infants, whose command of representational [43–45], lexical [46], and communicative capacities [5,47] suggests that they may be able to successfully resolve abstract verbal reference. By at least by 12–13 months, naming profoundly influences infants’ learning. Applied consistently, words help infants form conceptual categories [45,48,49], as well as guide their attention to commonalities among and differences between individual exemplars [50]. Further, infants as young as 13-months-old rely on labels to infer non-visible properties and extend them to novel instances of a category, even when they are not especially perceptually similar to the learning exemplars [e.g., 51,52]. These early capacities lead us to hypothesize that provided with a large enough lexicon, 12-month-olds may successfully learn novel words from language input alone. Experiment 2 investigates this possibility.

Focusing on 12-month-olds, however, raises a new challenge. For our semantic priming manipulation to succeed, infants must comprehend at least 3 words from each semantic neighborhood under investigation (e.g., fruits, clothing, vehicles). Yet 12-month-olds’ existing lexicons are rather sparse: only 30% of 12-month-olds meet this lexical requirement [28]. To redress this limitation, we initiate an at-home vocabulary training phase designed to bolster infants’ lexical comprehension before they participate in the experiment proper.

Method

Participants

Participants will be 72 infants (N = 24 per condition: Semantic Priming, No Priming, Follow-up Control) recruited from Chicagoland area. All recruitment and consent procedures will be identical to Experiment 1. Participants may be excluded and replaced due to fussiness, crying, or equipment failure.

Apparatus, stimuli, and data preparation will be identical to Experiment 1. All IVs, DVs, and analyses will also be the same as in Experiment 1, except that infants’ LT to the target on the familiar word knowledge test will be added to the main analyses.

Timeline

We intend to begin data collection as soon as the current registered report is granted in-principle acceptance. We expect that all data will be collected and analyzed within a calendar year from the completion of Experiment 1. The projected timeline to complete both proposed experiments is two calendar years from the in-principle acceptance date.

Procedure

The experimental procedure will be identical to Experiment 1 with one exception: before infants participate in the experiment proper, their parents will be enlisted to provide an at-home vocabulary training phase to bolster their infants’ lexical comprehension. Two weeks before their lab visit (or online participation on Lookit), participants’ parents will receive an online digital picture book containing 24 word-image pairs, including the 12 pairs featured in the Priming phase of the experiment proper. The order of word-image pairs will be randomized among participants. Parents will be asked to read their book at least once per day for 14 days, pointing to each picture and articulating its name. Before parents begin reading the book to their infants, they will view a video demonstration that will provide guidelines to reading this book to their infants and suggestions for establishing a daily reading routine (e.g., read before bedtime or after the nap, etc.). Moreover, parents will be contacted (email or text, as the parent prefers) every other day, reminding them to read the book to their infant. Parents will also be asked to fill out an online form daily (see S4 Appendix), reporting how engaged their infant was during the reading sessions. This at-home training should bolster infants’ lexical knowledge, permitting us to test their ability to learn words from displaced reference when they visit the lab.

When infants visit the lab (or participate in the study online via Lookit), we first test comprehension of the 12 words included in the experiment proper. On each of 12 trials, infants will see two images from their book—one that will be used in Priming (e.g., apple, a target) and another (e.g., spatula, a distractor). Following Bergelson and Swingley [2012; 53], infants will be prompted to look at the target (“Look, an apple!”). The proportion of LT to the target will be evaluated against chance (50%). It is possible that infants will fail to learn words used in Priming during the two-week exposure to picture-book reading. If this is the case, we will explore alternative methods of boosting infants’ word knowledge and consider alternative way of testing their ability to establish new word-object mappings from abstract reference.

Predictions

We expect that on the comprehension task that tests infants’ knowledge of the familiar 12 words, infants will exhibit excellent performance and look to the target images over 70% of the time over the course of the trial. In the experiment proper, if 12-month-olds can indeed establish a representation of word meaning from abstract verbal reference, they should successfully map the novel word to the unseen object, echoing the results of Experiment 1. We expect to observe a significant effect of Condition on both DVs–the overall proportion of LT to the target across 6 s and the time course of infants’ LT. We predict that the direction of differences will be the same as in Experiment 1. Further, if our hypothesis is correct, then infants’ performance on the comprehension test will be correlated with their success in the experiment proper. However, if despite vocabulary training, 12-month-olds fail to learn novel words, this will either suggest that these infants cannot yet resolve abstract verbal reference, or that their sensitivity to semantic priming is insufficiently robust to guide inferences about absent objects. Additional investigations will be required to distinguish between these possibilities.

Results and discussion

To be added when data collection and analyses are complete.

General discussion

Our goal of this paper was to fill the gap about the development of abstract verbal reference and address the outstanding questions about younger infants’ ability to learn novel words from language input alone. To be expanded when data collection and analyses are complete. Implications for the co-development of language and cognition and the origins of the human ability to extract information from symbolic input will be discussed.

Supporting information

S1 Appendix. Augmented MacArthur short form vocabulary checklist: Level II.

(PDF)

Click here for additional data file. (491.5KB, pdf)
S2 Appendix. Demographic questionnaire.

(PDF)

Click here for additional data file. (90.5KB, pdf)
S3 Appendix. Visual* and auditory stimuli in Experiments 1 and 2.

*Source of images in the Test phase: https://unsplash.com.

(PDF)

Click here for additional data file. (2.7MB, pdf)
S4 Appendix. Vocabulary training tracking form.

(PDF)

Click here for additional data file. (63.9KB, pdf)

Acknowledgments

We thank Caitlin Draper, Courtney Goldenberg, Rachel Kritzig, Ren Mondesir, Kiki Ogbuefi, Mary Okematti, Katelyn Pass, Judith Roeder, Annalisa Romaneneko, Murielle Standley, and Victoria Vizzini for help with pilot data collection. We also thank Miriam Novack and Alexander LaTourrette for their contribution to stimuli creation, helpful discussion about the study design, and help with pilot data analyses.

Data Availability

We agree to make our raw data, any digital study materials, and analysis code available on the Open Science Framework (https://osf.io/).

Funding Statement

Pilot data collection was funded by a National Institutes of Health grant #R01HD083310 awarded to SW. Any future funding sources that will be used to carry out the proposed research will be disclosed and acknowledged at the moment of the full manuscript submission. The funders had and will not have a role in study design, data collection and analysis, decision to publish, or preparation of the manuscript

References

  • 1.Deacon T. The symbolic species. New York: W; 1997. [Google Scholar]
  • 2.Clark HH, Marshall CR. Definite reference and mutual knowledge In: Elements of Discourse Understanding. edited by Joshi Aravind et al. Cambridge, UK: Cambridge University Press; 1981. p. 10–63. [Google Scholar]
  • 3.Miller GA. The Place of Language in a Scientific Psychology. Psychol Sci. 1990. January;1(1):7–14. [Google Scholar]
  • 4.Waxman SR, Gelman SA. Early word-learning entails reference, not merely associations. Trends in Cognitive Sciences. 2009. June;13(6):258–63. 10.1016/j.tics.2009.03.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Gliga T, Csibra G. One-year-old infants appreciate the referential nature of deictic gestures and words. Psychol Sci. 2009. March;20(3):347–53. 10.1111/j.1467-9280.2009.02295.x [DOI] [PubMed] [Google Scholar]
  • 6.Novack MA, Waxman S. Becoming human: human infants link language and cognition, but what about the other great apes? Phil Trans R Soc B. 2020. January 6;375(1789):20180408 10.1098/rstb.2018.0408 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Behne T, Liszkowski U, Carpenter M, Tomasello M. Twelve-month-olds’ comprehension and production of pointing: Twelve-month-olds comprehend pointing. British Journal of Developmental Psychology. 2012. September;30(3):359–75. 10.1111/j.2044-835X.2011.02043.x [DOI] [PubMed] [Google Scholar]
  • 8.Bohn M, Zimmermann L, Call J, Tomasello M. The social-cognitive basis of infants’ reference to absent entities. Cognition. 2018. August;177:41–8. 10.1016/j.cognition.2018.03.024 [DOI] [PubMed] [Google Scholar]
  • 9.Gräfenhain M, Behne T, Carpenter M, Tomasello M. One-year-olds’ understanding of nonverbal gestures directed to a third person. Cognitive Development. 2009. January;24(1):23–33. [Google Scholar]
  • 10.Rüther J, Liszkowski U. Ontogenetic Emergence of Cognitive Reference Comprehension. Cogn Sci [Internet]. 2020. July [cited 2020 Jul 16];44(7). Available from: https://onlinelibrary.wiley.com/doi/abs/10.1111/cogs.12869 [DOI] [PubMed] [Google Scholar]
  • 11.Tomasello M, Carpenter M, Liszkowski U. A New Look at Infant Pointing. Child Development. 2007. May;78(3):705–22. 10.1111/j.1467-8624.2007.01025.x [DOI] [PubMed] [Google Scholar]
  • 12.Liszkowski U, Carpenter M, Tomasello M. Reference and attitude in infant pointing. J Child Lang. 2007. February;34(1):1–20. 10.1017/s0305000906007689 [DOI] [PubMed] [Google Scholar]
  • 13.Gallerani CM, Saylor MM, Adwar S. Mother–Infant Conversation About Absent Things. Language Learning and Development. 2009. September 17;5(4):282–93. [Google Scholar]
  • 14.Osina MA, Saylor MM, Ganea PA. When familiar is not better: 12-month-old infants respond to talk about absent objects. Developmental Psychology. 2013;49(1):138–45. 10.1037/a0027903 [DOI] [PubMed] [Google Scholar]
  • 15.Osina MA, Saylor MM, Ganea PA. Object Locations, Identity and Absent Reference Understanding at 12 Months. Infancy. 2014. January;19(1):65–81. [Google Scholar]
  • 16.Saylor MM, Ganea P. Infants interpret ambiguous requests for absent objects. Developmental Psychology. 2007;43(3):696–704. 10.1037/0012-1649.43.3.696 [DOI] [PubMed] [Google Scholar]
  • 17.Bauer PJ, Wenner JA, Dropik PL, Werwerka SS. Parameters of Remembering and Forgetting in the Transition from Infancy to Early Childhood. Monographs of the Society for Research in Child Development. 2000;65(4):i-vi+1–213. [PubMed] [Google Scholar]
  • 18.Saylor MM. Twelve- and 16-month-old infants recognize properties of mentioned absent things. Developmental Sci. 2004. November;7(5):599–611. 10.1111/j.1467-7687.2004.00383.x [DOI] [PubMed] [Google Scholar]
  • 19.Saylor MM, Baldwin DA. Discussing those not present: comprehension of references to absent caregivers. J Child Lang. 2004. August;31(3):537–60. 10.1017/s0305000904006282 [DOI] [PubMed] [Google Scholar]
  • 20.Ganea PA. Contextual Factors Affect Absent Reference Comprehension in 14-Month-Olds. Child Development. 2005. September;76(5):989–98. 10.1111/j.1467-8624.2005.00892.x [DOI] [PubMed] [Google Scholar]
  • 21.Ganea PA, Fitch A, Harris PL, Kaldy Z. Sixteen-month-olds can use language to update their expectations about the visual world. Journal of Experimental Child Psychology. 2016. November;151:65–76. 10.1016/j.jecp.2015.12.005 [DOI] [PubMed] [Google Scholar]
  • 22.Luchkina E, Xu F, Sobel D, Morgan J. Sixteen-month-olds comprehend unanchored absent reference [Internet]. Open Science Framework; 2020. February [cited 2020 Mar 4]. Available from: https://osf.io/5tc6d. [Google Scholar]
  • 23.Arunachalam S, Syrett K. Specifying event reference in verb learning. 2014. [cited 2019 Dec 11]; Available from: https://rucore.libraries.rutgers.edu/rutgers-lib/48831/. [Google Scholar]
  • 24.Arunachalam S, Waxman SR. Meaning from syntax: Evidence from 2-year-olds. Cognition. 2010. March;114(3):442–6. 10.1016/j.cognition.2009.10.015 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Ferguson B, Graf E, Waxman SR. When Veps Cry: Two-Year-Olds Efficiently Learn Novel Words from Linguistic Contexts Alone. Language Learning and Development. 2018. January 2;14(1):1–12. 10.1080/15475441.2017.1311260 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Yuan S, Fisher C. “Really? She blicked the baby?”: two-year-olds learn combinatorial facts about verbs by listening. Psychol Sci. 2009. May;20(5):619–26. 10.1111/j.1467-9280.2009.02341.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Ferguson B, Graf E, Waxman SR. Infants use known verbs to learn novel nouns: Evidence from 15- and 19-month-olds. Cognition. 2014. April;131(1):139–46. 10.1016/j.cognition.2013.12.014 [DOI] [PubMed] [Google Scholar]
  • 28.Frank MC, Braginsky M, Yurovsky D, Marchman VA. Wordbank: an open repository for developmental vocabulary data. J Child Lang. 2017. May;44(3):677–94. 10.1017/S0305000916000209 [DOI] [PubMed] [Google Scholar]
  • 29.Bergelson E, Aslin R. Semantic Specificity in One-Year-Olds’ Word Comprehension. Language Learning and Development. 2017. October 2;13(4):481–501. 10.1080/15475441.2017.1324308 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Delle Luche C, Durrant S, Floccia C, Plunkett K. Implicit meaning in 18-month-old toddlers. Dev Sci. 2014. November;17(6):948–55. 10.1111/desc.12164 [DOI] [PubMed] [Google Scholar]
  • 31.Hendrickson K, Sundara M. Fourteen-month-olds’ decontextualized understanding of words for absent objects. J Child Lang. 2017. January;44(1):239–54. 10.1017/S0305000915000756 [DOI] [PubMed] [Google Scholar]
  • 32.Fenson L, Pethick S, Renda C, Cox JL, Dale PS, Reznick JS. Short-form versions of the MacArthur Communicative Development Inventories. Applied Psycholinguistics. 2000. March;21(1):95–116. [Google Scholar]
  • 33.Faul F, Erdfelder E, Lang A-G, Buchner A. G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods. 2007. May;39(2):175–91. 10.3758/bf03193146 [DOI] [PubMed] [Google Scholar]
  • 34.Cohen J. Statistical Power Analysis for the Behavioral Sciences [Internet]. 2nd ed. Routledge; 2013. [cited 2020 Sep 24]. Available from: https://www.taylorfrancis.com/books/9780203771587. [Google Scholar]
  • 35.Scott K, Schulz L. Lookit (Part 1): A New Online Platform for Developmental Research. Open Mind. 2017. February;1(1):4–14. [Google Scholar]
  • 36.Dale PS, Fenson L. Lexical development norms for young children. Behavior Research Methods, Instruments, & Computers. 1996. March;28(1):125–7. [Google Scholar]
  • 37.Arias-Trejo N, Plunkett K. Lexical–semantic priming effects during infancy. Phil Trans R Soc B. 2009. December 27;364(1536):3633–47. 10.1098/rstb.2009.0146 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Mani N, Plunkett K. In the Infant’s Mind’s Ear: Evidence for Implicit Naming in 18-Month-Olds. Psychol Sci. 2010. July;21(7):908–13. 10.1177/0956797610373371 [DOI] [PubMed] [Google Scholar]
  • 39.Wojcik EH. The Development of Lexical-Semantic Networks in Infants and Toddlers. Child Dev Perspect. 2018. March;12(1):34–8. [Google Scholar]
  • 40.Dink JW, Ferguson B. eyetrackingR: An R Library for Eye-tracking Data Analysis [Internet]. 2015. Available from: http://www.eyetrackingr.com. [Google Scholar]
  • 41.Csibra G, Volein Á. Infants can infer the presence of hidden objects from referential gaze information. British Journal of Developmental Psychology. 2008. March;26(1):1–11. [Google Scholar]
  • 42.Martin A, Onishi KH, Vouloumanos A. Understanding the abstract role of speech in communication at 12months. Cognition. 2012. April;123(1):50–60. 10.1016/j.cognition.2011.12.003 [DOI] [PubMed] [Google Scholar]
  • 43.Vouloumanos A, Onishi Kristin H, Pogue A. Twelve-month-old infants recognize that speech can communicate unobservable intentions. Proceedings of the National Academy of Sciences. 2012. August 7;109(32):12933–7. 10.1073/pnas.1121057109 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Vouloumanos A, Waxman SR. Listen up! Speech is for thinking during infancy. Trends in Cognitive Sciences. 2014. December;18(12):642–6. 10.1016/j.tics.2014.10.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Waxman SR, Markow DB. Words as Invitations to Form Categories: Evidence from 12- to 13-Month-Old Infants. Cognitive Psychology. 1995. December;29(3):257–302. 10.1006/cogp.1995.1016 [DOI] [PubMed] [Google Scholar]
  • 46.Fenson L, Dale PS, Reznick JS, Bates E, Thal DJ, Pethick SJ, et al. Variability in Early Communicative Development. Monographs of the Society for Research in Child Development. 1994;59(5):i [PubMed] [Google Scholar]
  • 47.Csibra G. Recognizing Communicative Intentions in Infancy. Mind & Language. 2010. March 15;25(2):141–68. [Google Scholar]
  • 48.Ferguson B, Waxman S. Linking language and categorization in infancy. J Child Lang. 2017. May;44(3):527–52. 10.1017/S0305000916000568 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Perszyk DR, Waxman SR. Linking Language and Cognition in Infancy. Annu Rev Psychol. 2018. January 4;69(1):231–50. 10.1146/annurev-psych-122216-011701 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.LaTourrette AS, Waxman SR. Naming guides how 12-month-old infants encode and remember objects. Proceedings of the National Academy of Sciences. 2020. 10.1073/pnas.2006608117 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Graham SA, Kilbreath CS, Welder AN. Thirteen-Month-Olds Rely on Shared Labels and Shape Similarity for Inductive Inferences. Child Development. 2004. March;75(2):409–27. 10.1111/j.1467-8624.2004.00683.x [DOI] [PubMed] [Google Scholar]
  • 52.Welder AN, Graham SA. The Influence of Shape Similarity and Shared Labels on Infants’ Inductive Inferences about Nonobvious Object Properties. Child Development. 2001. November;72(6):1653–73. 10.1111/1467-8624.00371 [DOI] [PubMed] [Google Scholar]
  • 53.Bergelson E, Swingley D. At 6–9 months, human infants know the meanings of many common nouns. Proceedings of the National Academy of Sciences. 2012. February 28;109(9):3253–8. 10.1073/pnas.1113380109 [DOI] [PMC free article] [PubMed] [Google Scholar]

Decision Letter 0

Lucas Payne Butler

10 Aug 2020

PONE-D-20-14424

Category priming promotes infants’ success in imagining and naming things unseen

PLOS ONE

Dear Dr. Luchkina,

First of all, thanks for your patience waiting for this decision. As I mentioned in a previous email, it was a challenge to find reviewers during this challenging time, and the timing of when those reviews came in delayed my decision.  After careful consideration, we feel that it your submission merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

I think this is an interesting proposal, and with some additional details and shifts in framing, as well as some possible modifications of some of the methodological details, could make a nice contribution to the literature on word learning. I won't go into detail with regard to the reviewers' comments, but let me highlight a few key points that I want to make sure you pay particular attention to:

1) Both reviewers thought that the theoretical motivation for these studies needed to be fleshed out, and had suggestions for how to do so. Reviewer 2 in particular was unclear what question this was asking that was distinct from prior work on word learning and learning about absent referents, and suggested that the work needed to be more clearly situated in the word learning literature. 

2) Both reviewers had some concerns about specific aspects of the methodology, including sample size, the possible need for an additional control condition, and more details about and justifications for specific methodological decisions.

3) Reviewer 2 notes, and I agree, that you may need to discuss how data collection will be impacted by the current pandemic. Is the plan to wait until after the pandemic subsides (which could be a long time), or are there plans for virtual data collection, in which case is there evidence that this will work with infants of this age? 

Please also be sure to read each review closely, as my above summary is by no means an exhaustive list of the recommended changes. 

Please submit your revised manuscript by Aug 29 2020 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

  • A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

  • A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

  • An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Lucas Payne Butler

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. There are several instances where placeholders have been inserted into the text, to be completed once results have been obtained, e.g. " To be expanded when data collection and analyses are complete.".

Please note that, should the article be accepted, the submitted Protocol will be published as an article that will be seperate from the final Research Article that includes results.

Please revise the text accordingly, keeping in mind that the Registered Report Protocol should be considered as a standalone article.

For further details please see here: https://journals.plos.org/plosone/s/other-article-types#loc-registered-reports.

3. We note that you have stated that you will provide repository information for your data at acceptance. Should your manuscript be accepted for publication, we will hold it until you provide the relevant accession numbers or DOIs necessary to access your data. If you wish to make changes to your Data Availability statement, please describe these changes in your cover letter and we will update your Data Availability statement to reflect the information you provide.

4. We note that Figures 1 and 2 in your submission contain copyrighted images.

All PLOS content is published under the Creative Commons Attribution License (CC BY 4.0), which means that the manuscript, images, and Supporting Information files will be freely available online, and any third party is permitted to access, download, copy, distribute, and use these materials in any way, even commercially, with proper attribution. For more information, see our copyright guidelines: http://journals.plos.org/plosone/s/licenses-and-copyright.

We require you to either (a) present written permission from the copyright holder to publish these figures specifically under the CC BY 4.0 license, or (b) remove the figures from your submission:

a. You may seek permission from the original copyright holder of Figures 1 and 2 to publish the content specifically under the CC BY 4.0 license.

We recommend that you contact the original copyright holder with the Content Permission Form (http://journals.plos.org/plosone/s/file?id=7c09/content-permission-form.pdf) and the following text:

“I request permission for the open-access journal PLOS ONE to publish XXX under the Creative Commons Attribution License (CCAL) CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). Please be aware that this license allows unrestricted use and distribution, even commercially, by third parties. Please reply and provide explicit written permission to publish XXX under a CC BY license and complete the attached form.”

Please upload the completed Content Permission Form or other proof of granted permissions as an "Other" file with your submission. 

In the figure caption of the copyrighted figure, please include the following text: “Reprinted from [ref] under a CC BY license, with permission from [name of publisher], original copyright [original copyright year].”

b.    If you are unable to obtain permission from the original copyright holder to publish these figures under the CC BY 4.0 license or if the copyright holder’s requirements are incompatible with the CC BY 4.0 license, please either i) remove the figure or ii) supply a replacement figure that complies with the CC BY 4.0 license. Please check copyright information on all replacement figures and update the figure caption with source information. If applicable, please specify in the figure caption text when a figure is similar but not identical to the original image and is therefore for illustrative purposes only.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Does the manuscript provide a valid rationale for the proposed study, with clearly identified and justified research questions?

The research question outlined is expected to address a valid academic problem or topic and contribute to the base of knowledge in the field.

Reviewer #1: Yes

Reviewer #2: Partly

**********

2. Is the protocol technically sound and planned in a manner that will lead to a meaningful outcome and allow testing the stated hypotheses?

The manuscript should describe the methods in sufficient detail to prevent undisclosed flexibility in the experimental procedure or analysis pipeline, including sufficient outcome-neutral conditions (e.g. necessary controls, absence of floor or ceiling effects) to test the proposed hypotheses and a statistical power analysis where applicable. As there may be aspects of the methodology and analysis which can only be refined once the work is undertaken, authors should outline potential assumptions and explicitly describe what aspects of the proposed analyses, if any, are exploratory.

Reviewer #1: Partly

Reviewer #2: Partly

**********

3. Is the methodology feasible and described in sufficient detail to allow the work to be replicable?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors described where all data underlying the findings will be made available when the study is complete?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception, at the time of publication. The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above and, if applicable, provide comments about issues authors must address before this protocol can be accepted for publication. You may also include additional comments for the author, including concerns about research or publication ethics.

You may also provide optional suggestions and comments to authors that they might find helpful in planning their study.

(Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This is a well written manuscript, with clear hypothesis. The design is clever, thorough and deals with potential confounds, thus allowing a straightforward validation of hypothesis. However, some additional clarifications are required and some additional confounds addressed.

The intro should discuss evidence for superordinate category knowledge in the second year of life. Your experiment depends on children being able to recognise a new fruit as belonging to the fruit category, for example. More detail needed also on the prerequisites for the semantic priming - The study you cite in support of such an ability (19) only shows that younger infants have broader categories, e.g. foot includes socks - it is not a study suggesting semantic priming. The developmental change described in that study might suggest that 15 mo do not even benefit from this over-generalization. Also, is priming sufficient, do infants not also need to make a pragmatic inference - that since the experimenter choose to name 3 fruits, the fourth must be from the same superordinate category? This is quite a peculiar inference which also needs to be backed up with some theoretical or empiric support from the literature. You preliminary study shows evidence that either semantic priming or pragmatic inferences do work in 15mo, so what is needed is some stronger apriori motivation of why it might be so.

Please provide power analysis for your sample - how many participants are needed to reach significance for both the main effect of Condition and the follow-up tests, for the expected (or medium) effect size. Please ensure that recruitment takes into account drop-out rate, thus ensuring that the final sample size is that expected from power analysis. Also, indicate the follow-up tests and any multiple correction applied, as part of your analysis.

Follow-up control condition - Looking towards the target after the first prompt is not strong evidence that they have not associated the word with the target - participants may make the pragmatic choice of looking at the category that the person engaged with previously. A better test would be to use a completely novel word in test.

Experiment 2- related to the comment above, if the training is aimed at teaching superordinate category knowledge, will labeling basic level categories achieve that ? If it is assumed 12 mo possess the superordinate categories, do they need to know the basic level labels in order to succeed at the task ?

Reviewer #2: Discussion of Theoretical Framing of the Research Question

Overall this is a well written proposal. I have several concerns related to the framing of the study and the specific research question.

The introduction starts by stating that (1) calling to mind referents in their absence is a crucial feature of human language, (2) this ability is enabled by an understanding of the referential understanding of words, and (3) the developmental origin of this referential understanding is not clear.

The authors then discuss some literature on bootstrapping on word meaning from syntax, and state that 15-month-olds fail to use syntax as a cue to a word’s meaning and pose the question of whether 15-month-olds have the conceptual and linguistic knowledge to learn words from language alone. Their study is aimed to answer this question.

I found this set-up of the study questions very confusing for the following reasons:

1. There is plenty of evidence in the word learning literature that infants have an understanding of the referential nature of labels by 15-months. For example, work by Bergelson and Swingley shows that 6-9 month olds have a fairly abstract representation of labels; several studies by Ganea and Saylor show that 12-15 month-olds respond to labels in a displaced reference context; Ganea, Fitch, Harris and Kaldy (2016) show that even 15-month-olds can update expectations about the visual word on the basis of language alone; and, directly relevant to this study’s goals, a recent study by Osina, Saylor, and Ganea (2018) indicates that infants as young as 16-months use category label knowledge to interpret absent reference. Also, the findings of Gliga and Csibra (2009) indicate that infants expect to find an objects based on a novel label coupled with a deictic gesture. Given this body of work (this is only a small selection of it) showing that infants have a referential understanding of language by 15 months, as shown by their ability to process labels in the absence of their referents and their ability to access categorical knowledge when hearing a label, it is not clear to me what the authors define as “referential understanding.” The authors need to state explicitly why these other ways of testing referential understanding do not provide sufficient evidence and in what way their study is advancing what we already know about infants’ referential understanding.

2. The study should be better situated in the word learning literature. It is not clear to me why the work on syntactical bootstrapping is included. This study is not using syntax but conceptual knowledge of a category to see whether children can infer meaning. Therefore a a review of the literature on the taxonomic bias on word learning needs to be included, rather than of the syntactical bootstrapping literature.

Given the above considerations, it was difficult to fully comprehend the exact research questions. It seems like the goal is to test whether toddlers establish a referent when they hear something even if they do not see the object, when it is taxonomically related to previous objects in the task. The novelty seems to be that a specific referent (using a novel not familiar word) can be recalled when the category is primed. What is the nature of children’s expectations of what a novel word means? Can children learn words in the absence of information, and does the same taxonomic bias occur without visual information?

Discussion of the Design

- For the condition effects, there were only predictions for referential understanding when toddlers were primed by category. There should also be expectations for the no category priming condition (the basis for this expectation also needs to be supported by the literature and then cited accordingly).

- Since the goal of the proposed research is to test the influence of conceptual priming on infants’ referential understanding, the perceptual cues should be minimized/held constant within and across conditions. Some perceptual features of the current stimuli might be potential confounding variable in the proposed study. For example, in the first trial of the Category-priming condition, the novel fruit is both conceptually and perceptually similar to the preceding set of stimuli while the non-target item (furniture) is both conceptually and perceptually distinct. An out-of-category item which perceptually looks more similar to the target category would work better.

- Decisions for how the question for the control condition was selected need to explained. Why and how they help address the research question.

- Why 6 seconds? Explain why this amount of time was selected and if there is research to suggest that this is the norm. Also, the exclusion criteria for the looking data needs to be clear. What is the threshold criteria to be included for each trial, and how was this decided.

- Explain the sample size – power analysis

- Why did the authors select this type of design as opposed to a within-subjects design? They could use 2 trials per condition instead of 4 – increases the power.

- What will the authors do with the information collected from the demographic questionnaire? Will they link any of the data to the study’s DVs?

- The timeline for data collection may need to be adjusted given COVID situation.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Teodora Gliga

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

PLoS One. 2021 Jan 7;16(1):e0244968. doi: 10.1371/journal.pone.0244968.r002

Author response to Decision Letter 0

28 Oct 2020

Dear Dr. Butler,

Please consider our revised manuscript PONE-D-20-14424 for publication in PLOS ONE. We are grateful to you and the reviewers for your thoughtful feedback. We hope that you agree that by responding to this very productive set of reviews, we have produced a significantly stronger manuscript. Please see below point-by-point responses to reviewer’s comments.

Sincerely,

Elena Luchkina

Summary of the major changes:

1. We clarified that the central construct under investigation is “abstract verbal reference”, which we define operationally as “an appreciation that words are linked to mental representations that can be established, retrieved and modified, even when the entities to which they refer are not perceptually available”. See §1 of the Introduction, as well as §2 about “displaced/fragmented reference” as its operationalized forms.

2. Following R1’s and R’s excellent suggestion, we no longer frame the work in the literature on superordinate category knowledge. We now focus on semantic neighborhoods as our main experimental manipulation. We have changed the title to reflect this modification and renamed the conditions from Category Priming and No-Category Priming conditions to Semantic Priming and No Priming conditions respectively.

3. Based on R1’s suggestion, we redesigned the Follow-up Control condition. It now addresses the possibility that success in the Semantic Priming condition is due to semantic relations among the objects themselves and is not dependent upon infants’ inferences about novel word meanings.

4. Per R2’s suggestion, we expanded the introduction to better situate the current project within the word-learning and reference literatures.

5. We also modified the test stimuli included in all experiments and conditions to address R2’s suggestion to minimize perceptual distance between target and distracter images.

Reviewer 1

The reviewer suggested that the test for our hypothesis would be stronger if we do not to locate our predictions in relation to superordinate category knowledge in infants this young.

We are grateful to R1 for the suggestion. We agree that we need not rely upon superordinate category knowledge to test abstract verbal reference. Instead, we frame the hypothesis in the context of the semantic priming available from known words. We propose that the ability to establish a representation of a hidden referent based on a semantic neighborhood would constitute strong evidence of resolving abstract reference (see p. 6, §1).

The reviewer also raised a concern that infants’ looking behavior in the Semantic Priming condition (formerly “Category Priming condition”) could be driven by the relations between the objects rather than by infants’ inferences about the meaning of the novel word.

We thank the reviewer for bringing this up. In response, we changed the design of the Follow-up Control condition, which now addresses this concern. Now, the Follow-up Control condition is designed identically to the Semantic Priming condition, except that infants hear an unfamiliar novel noun during test trials (see p. 10, §1). This way, we can discern between the effects of the relation between the objects and infants’ semantic inferences about the novel word.

The reviewer asked that we provide power analysis for our sample, taking into account the drop-out rate, and indicate what multiple corrections would be applied.

We now added a power analysis and a recruitment plan that incorporates the projected drop-out rate (p. 6, §2). Because our planned analyses are based on GLMM, which (as we now point out) includes Test Item as a random factor, we do not need additional multiple corrections.

The reviewer suggested that we use a completely novel word during Test in the Follow-up Control condition to evaluate whether infants associated the word with the target object.

As mentioned above, we have done so. (p. 12, §1).

Reviewer 2

The reviewer asked that we ground our work more comprehensively in the literature on word learning and outline the new insights that our work will bring to the problem of verbal reference.

We appreciate this suggestion and have done so, now including the references cited by the reviewer and many others (pp. 3-4). We highlighted the difference between the extant empirical investigations and the current proposal, as well as our contribution to this literature (p. 4, §2; “What remains unknown is when infants successfully acquire new representations and update them on the basis of language alone.”).

The reviewer also suggested we define more clearly the construct under investigation.

We appreciate this suggestion. We clarified that the central construct under investigation is “abstract verbal reference”, which we define on p. 1, §1.

The reviewer asked that we explicitly state our predictions for each condition.

We have now added explicit predictions for each condition (pp. 12-13).

The reviewer suggested that we minimize the perceptual differences between test images within and across conditions.

We appreciate this suggestion, and have edited the test stimuli accordingly (see Appendix 3).

The reviewer asked us to clarify the advantages of the Follow-up Control condition, pointing out how they help address the research question.

We appreciate this suggestion. In response, we have clarified that the Follow-up Control condition (which now has been modified following R1’s suggestion; see p. 10, §1) will help discern between the effects of the relation between the objects and infants’ semantic inferences about the novel word.

The reviewer suggested that we clarify how we decided on the duration of test trials and that we outline exclusion criteria for the looking data.

We now articulate the reasons for 6 s test trials and the threshold criteria in the pilot and proposed studies. Both are based on earlier investigations of infants’ ability to take advantage of existing word knowledge to infer the meaning of novel words (p. 10, § 2; ).

The reviewer asked that we explain the sample size and provide power analysis.

We now added power analysis on p. 6, §2.

The reviewer asked that we clarify why we chose to a within-subjects design with 4 test trials.

We have done so on p. 8, §2.“We selected a between-subject design to avoid potential carry-over effects, which would dilute the predicted advantage of Semantic Priming”.

The reviewer also asked us to explain to we would analyze the information collected from the demographic questionnaire and whether it would be linked to the DVs.

The demographic data will used in preliminary analyses and may be included in the main analyses if the preliminary analyses show a significant effect. We do not have specific hypotheses about the role of the demographic factors in the development of abstract verbal reference comprehension. We clarify this on p. 12, §2.

The reviewer suggested that we provide a COVID-19 contingency plan and outline an adjusted timeline for data collection.

We added a COVID-19 contingency plan and explained the projected timeline of data collection on p. 7, §2.

Attachment

Submitted filename: Response to Reviewers.docx

Click here for additional data file. (19.9KB, docx)

Decision Letter 1

Lucas Payne Butler

21 Dec 2020

Semantic priming supports infants’ ability to learn names of unseen objects

PONE-D-20-14424R1

Dear Dr. Luchkina,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Lucas Payne Butler

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Does the manuscript provide a valid rationale for the proposed study, with clearly identified and justified research questions?

The research question outlined is expected to address a valid academic problem or topic and contribute to the base of knowledge in the field.

Reviewer #2: Yes

**********

2. Is the protocol technically sound and planned in a manner that will lead to a meaningful outcome and allow testing the stated hypotheses?

The manuscript should describe the methods in sufficient detail to prevent undisclosed flexibility in the experimental procedure or analysis pipeline, including sufficient outcome-neutral conditions (e.g. necessary controls, absence of floor or ceiling effects) to test the proposed hypotheses and a statistical power analysis where applicable. As there may be aspects of the methodology and analysis which can only be refined once the work is undertaken, authors should outline potential assumptions and explicitly describe what aspects of the proposed analyses, if any, are exploratory.

Reviewer #2: Yes

**********

3. Is the methodology feasible and described in sufficient detail to allow the work to be replicable?

Reviewer #2: Yes

**********

4. Have the authors described where all data underlying the findings will be made available when the study is complete?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception, at the time of publication. The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above and, if applicable, provide comments about issues authors must address before this protocol can be accepted for publication. You may also include additional comments for the author, including concerns about research or publication ethics.

You may also provide optional suggestions and comments to authors that they might find helpful in planning their study.

(Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #2: The authors have responded to all my suggestions. I look forward to seeing the results. Upon writing the paper I suggest the authors also cite this relevant work:

Saylor, M. M., Osina, M., Tassin, T., Rose, R., & Ganea, P. A. (2016). Creature feature: preschoolers use verbal descriptions to identify referents. Journal of Experimental Child Psychology. doi:10.1016/j.jecp.2016.07.005

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #2: No

Acceptance letter

Lucas Payne Butler

23 Dec 2020

PONE-D-20-14424R1

Semantic priming supports infants’ ability to learn names of unseen objects

Dear Dr. Luchkina:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Lucas Payne Butler

Academic Editor

PLOS ONE

Associated Data

    This section collects any data citations, data availability statements, or supplementary materials included in this article.

    Supplementary Materials

    S1 Appendix. Augmented MacArthur short form vocabulary checklist: Level II.

    (PDF)

    Click here for additional data file. (491.5KB, pdf)
    S2 Appendix. Demographic questionnaire.

    (PDF)

    Click here for additional data file. (90.5KB, pdf)
    S3 Appendix. Visual* and auditory stimuli in Experiments 1 and 2.

    *Source of images in the Test phase: https://unsplash.com.

    (PDF)

    Click here for additional data file. (2.7MB, pdf)
    S4 Appendix. Vocabulary training tracking form.

    (PDF)

    Click here for additional data file. (63.9KB, pdf)
    Attachment

    Submitted filename: Response to Reviewers.docx

    Click here for additional data file. (19.9KB, docx)

    Data Availability Statement

    We agree to make our raw data, any digital study materials, and analysis code available on the Open Science Framework (https://osf.io/).

    Articles from PLoS ONE are provided here courtesy of PLOS

    RESOURCES