The meaning of "that": recency and the importance of context
« previous post | next post »
"What Does 'that' Mean in 'Did You See That?'" ScienceDaily, March 25, 2025. Discussing "Recency and Rarity Effects in Disambiguating the Focus of Utterance: A Developmental Study." Kishimoto, Reiki et al. PLOS ONE 20, no. 2 (February 12, 2025): e0317433.
Summary (of the Science Daily article)
What goes through your mind when someone asks 'Did you see that?' A study shows that children and adults process ambiguous information like 'that' differently. While children focus more on the literal meaning, adults use multiple cues to grasp the speaker's intent. Researchers hope their findings could aid dialogue robot development and improve support systems for individuals with communication difficulties.
Second paragraph of the Science Daily article
Imagine watching a quiet night sky. When a shooting star streaks across the sky and someone asks, "Did you see that?," we naturally understand they are referring to the shooting star, not the twinkling stars in the background.
A striking instance of the importance of context for communication. Adults are better than children at sorting it out.
Abstract (of the PLOS ONE article)
While the communicator’s intended referent in a conversation may not be immediately apparent, effective communication often overcomes this ambiguity. However, the specific mechanisms through which children use various cues to pinpoint the referent remain unclear. The communicator determines what is salient from the receiver’s perspective. In return, the receiver identifies what the communicator identifies to be relevant for the receiver. The current study focused on two salient cues: rarity and recency, because rarity results in surprise and recency means a cue is more easily perceived and remembered. The current study investigated how adults and children aged 7–10 employ rarity and recency cues embedded in a series of events to clarify the referent intended by the communicator. Participants observed sequences comprising one rare and eight frequent events. An utterance, “Did you see that?” was presented at the end of each sequence, and participants identified the event(s) referred to by “that.” Events that were rare and close to the utterance were more likely to be identified as the referents. Notably, the utilization of these cues differed between adults and children.
For adults, the recency effect manifested gradually, with events closer to the utterance identified more frequently, and it exhibited an interaction with rarity. Among children, the recency effect was absolute, as the event closest to the utterance held a higher likelihood of being identified, and this effect was not influenced by rarity. Two additional conditions eliminated potential response biases and memory-related confounds. Our research suggests that school-age children are capable of disambiguating utterances by factoring in that the events they find salient are likely to be the communicator’s focus. However, they are still in the process of developing reasoning skills similar to those of adults.
This kind of linguistic recency is quite different from the "recency illusion" that was a regular topic of discussion in the early years of Language Log. The latter kind of recency was often coupled with "frequency", whereas the recency discussed in the work of Kishimoto et al. is paired with rarity, just the opposite vector.
Selected readings
- "This 'n that" (12/13/24)
- "Baby talk, part 2" *8/19/18)
[Thanks to Ted McClure]
Chris Button said,
April 2, 2025 @ 6:48 pm
I haven't read the article (so probably shouldn't comment), but I note the Japanese name of the main author. I wonder therefore if distinctions like Japanese sore "that" and are "that over there" are considered in terms of how geographically removed the context is from the location of the interlocutors.
Lucas Christopoulos said,
April 2, 2025 @ 7:10 pm
@Chris
How would you translate それか な (なななななな) in a geographical and temporal context? uncertainty ?
Lucas Christopoulos said,
April 2, 2025 @ 7:14 pm
I meant the な naaaaaaaaaaaaaaa
Jonathan Smith said,
April 2, 2025 @ 7:15 pm
The authors are Japanese / the study was conducted on Japanese speakers / the operative phrase was Ima-no mita (= did [you] see [it / what happened] just now?) I.e. the word "that" or a direct parallel wasn't involved. So problems in Science Daily / LL write-ups. I guess many "pro-drop"-ish languages will look similar here — cf. Chinese "你[剛才]看見了嗎"? or some such.
Chris Button said,
April 2, 2025 @ 7:23 pm
@ Lucas Christopoulos
You mean それかな…?
Chris Button said,
April 2, 2025 @ 8:12 pm
今の見た? (Ima no mita?): "Did you see [that just] now"
それを見た? (Sore o mita?): "Did you see that?"
Lucas Christopoulos said,
April 2, 2025 @ 9:36 pm
@Chris Button
Yes そかな…with a long aaaaaaaaaa at the end. I guess sooooo? I think sooooo? It is not a real agreement though, but a half agreement?