Talk:Subitizing

Pronunciation
Hi. At a recent conference, I had a discussion about how to correctly pronounce the word 'subitize'. Is it 'soobitize' or 'subitize' (as in 'subtract')?--Jo-Anne LeFevre 01:48, 18 July 2005 (UTC)

Its 'soobitize' as far as I know patrickw


 * I am American and have only heard "soobitize" (IPA:/'subɪtaɪz/). But I notice that all British sources I have found on the Internet say "subbitize" /'sʌbɪtaɪz/. --seberle (talk) 05:01, 17 November 2020 (UTC)

Citation
Near the beginning of the 'Imaging Enumeration' subsection there is a 'citation needed' tag where there used to be the text "(Sathian et al., 1999)"; it would be nice if someone would figure out what this is and refer to it correctly. Archelon 23:05, 2 December 2006 (UTC)


 * The reference is:


 * Sathian, K., Simon, T.J., Peterson, S., Patel, G.A., Hoffman, J.M., & Grafton, S.T. (1999). Neural evidence linking visual object enumeration and attention. Journal of Cognitive Neuroscience, 11(1), 36-51. patrickw 06:59, 23 April 2007 (UTC)

Subitizing

 * I'm not sure why this article is titled "Subitizing and counting". It seems to be exclusively about subitizing.  Rename?  70.247.160.102 (talk) 06:11, 21 February 2010 (UTC)

Move?
Subitizing and counting → — User:70.247.169.94 02:48, 4 July 2010 (UTC)

Sounds good to me. This article is about subitizing, and discusses counting only for contrast. — Gavia immer (talk) 18:09, 4 July 2010 (UTC) billinghurst  sDrewth  05:05, 14 July 2010 (UTC)
 * moved

Educational applications Section
The whole chapter is quite bad. It used to be called Mathematical Applications. Maybe it should simply be called Applications?

My attention was first drawn to the chapter because it claims that "Some practitioners claim that with training, children are capable of subitizing 15+ objects correctly" and that claim is backed up by a link to a single commercial early education service. Furthermore the backing for the claim was not instantly obvious at the linked site. And in any case its quite bold to claim that more than one practitioners are of an opinion based on one example. In my opinion we either site more and better examples or remove the claim.

--Juhoeemeli (talk) 07:30, 7 April 2014 (UTC)

The educational use of subitizing training in dyslectic children has been thoroughly studied scientifically by the group of the late Prof. Burkhart Fischer in Freiburg. That should be added sometime. Strasburger (talk) 10:59, 5 April 2019 (UTC)

Chimps, bonobos and other aniimals
I saw a documentary film on counting abilities in animals (possibly on PBS - Nova perhaps?) and it showed how quickly chimps and bonobos could count objects. While humans are good at up to 5 or so objects, these animals were faster than humans when 6-10 objects were presented. I wonder if a section on animal abilities would be right for this article?99.245.230.104 (talk) 19:12, 21 April 2014 (UTC)


 * It came on my mind too. Extermely enhanced subitizing abilities of some savants should be mentioned as well. —Mykhal (talk) 09:01, 25 April 2017 (UTC)

"Numbers 1-10" image inexplicably in "Enumerating Afterimages" section
There is currently an image labeled "numbers 1-10" in the "Enumerating afterimages" section. This image is interesting but I cannot tell what the significance of it is and it certainly does not seem to be related to the enumeration of afterimages. What is the intention of this image and is there a better place for it in the article / something that can be said about it? I am tempted to remove it, but it *is* a good illustration of dividing large sets into subitizable groups even though it's out of context there. --24.185.86.218 (talk) 14:39, 18 July 2014 (UTC)


 * I have moved this image into the main section from the "Enumerating afterimages" section. It's a bit awkward in terms of formatting but at least it's not in the wrong place any more. --24.185.86.218 (talk) 01:30, 17 September 2014 (UTC)


 * Given that the lede already has an image of dividing ten stars into four groups, this 1-10 image seems redundant. I've cut it. I can't see that the article actually says anything about dividing large sets into subitizable groups, though? --McGeddon (talk) 11:33, 9 May 2016 (UTC)

Mathematical logic category?
I am puzzled to see that this article is included in the category "Mathematical logic". Mathematical logic is a rather abstruse field concerned with hierarchies of sets, infinite cardinals, consistency strengths and the like, not such lowly things as counting a handful of objects. But perhaps there's a connection I'm missing. 209.179.87.11 (talk) 06:13, 3 August 2016 (UTC)

The claim that toddlers can subitize to 100
I read Doman's 1982 "Teach Your Baby Math" where he claimed https://books.google.com/books?id=3nLsDYOi-rsC&focus=searchwithinvolume&q=dots that a toddler could distinguish (subitize) 38 dots from 39, and 91 from 92, without counting, up to 100.

The "Institutes for the Achievement of Human Potential" BrillKids' 2008 book makes the same claim http://www.brillkids.com/media/ebooks/ebook-teaching-your-baby-math.pdf "When a very young child looks at a large number of items, she doesn’t need to guess or count to see how many are there. The child can instantly see 48 marbles, in just the same way that the rest of us can instantly see 4- by subitizing. ... Once the child has a foundation in quantity (having learned up to at least the number 20), you can begin equations using those quantities (while continuing to teach up to 100)." BUT then they claim, "The ability to subitize large quantities will typically fade away as a child grows older... children retain the ability to perceive large quantities at least up to the age of two and a half."

From reading Doman's 1982 book decades ago, and now using Google Books to search, I don't think he indicated that the children lose this alleged ability.

Question: Is this "Infant Subitizing" worth a subsection under Educational_applications (with criticism) or perhaps in a new section titled Controversies? Bob Enyart, Denver KGOV radio host (talk) 17:46, 10 July 2017 (UTC)

Terms
I think we need to be careful in the opening with our terms. Currently the third sentence in the lead paragraph is the following:
 * Number judgments for larger set-sizes are referred to either as counting if sufficient time is available or as estimating if it is not.

This sentence has gone through several edits recently, including one half-hearted edit of my own which wasn't very good. None of them are very satisfactory. I think we should simply delete this sentence. It is more confusing than helpful.

To be clear: Subitizing is a psychological phenomenon whereby one apprehends at once the number of items. It is almost always used for small numbers of items. Counting generally refers to a complex skill, typically learned at a young age, where one establishes a bijection between the items in a set and the counting numbers. One can count both small and large sets; we are not obligated to subitize small sets.

There is a related skill of estimation whereby an approximate number is assigned to a set without counting. Last I checked (correct me if some recent research has been done), most research indicated that estimation and subitizing are probably two distinct skills.

Currently the third sentence implies that one does not "count" small sets, and perhaps even that the terminology depends on the amount of time available. Estimation is, of course, faster than counting. But someone might choose to estimate even if time is available. The action is not defined by the time available, but rather by the procedure used.

If we wanted to salvage the last sentence, we need to be clear that there are three known methods of quantifying a set (subitizing, counting, estimation) and that these are defined by the distinct method used. (They are not defined by the time available, and only partially by the size of the set).

But I recommend we simply delete the third sentence. Such details can be discussed elsewhere in the article. --seberle (talk) 08:01, 24 March 2021 (UTC)


 * Since the sentence is unsourced anyway, I would agree it is a good idea to remove it. Strasburger (talk) 08:49, 10 May 2021 (UTC)
 * For the record: I changed "time available" to "mental effort invested" -- without having seens this discussion. I think the statement is true, as such, or at least closer to being true now, but yes, it is still unsourced. I referred to Thinking, Fast and Slow in my edit summary, but I don't think Daniel Kahneman actually mentions subitizing (the word is not in the index).--Nø (talk) 09:13, 10 May 2021 (UTC)
 * Hmm. That's better, for sure. But it is still not completely accurate. Subitizing is easier, yes. But it's really the method used that defines what subitizing is, not the effort. The method is often inferred in clinicial trials by the amount of time the person takes. But that doesn't mean subitizing is defined by time or effort. If I have time, I'll try to come up with something more accurate. --seberle (talk) 01:39, 10 June 2021 (UTC)

Dominos, and fixed patterns
The section "Other applications" says:
 * Dice, playing cards and other gaming devices traditionally split quantities into subitizable groups with recognizable patterns.

Apart from possibly adding dominos to the list of games, I'd like to say that I think using recognizable patterns invokes a different mental mechanism from what is in general called subitizing. It's pattern recognition, which unlike subitizing dependents heavily on the arrangement of the elements. Introspectively, for me, the "five" (a quincunx) and "six" on a die, the "nine" on double-nine dominos, and a pattern of 7 dots in a regular hexagon with one dot in the centre, are clear examples of this. They are recognized (and associated with a number) much more easily than objects in a random arrangement, and I doubt it is by the same mechanism. The actual splitting into subitizable groups referred to in in the quoted sentence is irrelevant; the only clear example of this I can readily find among the games listed is the "seven" of playing cards, arranged with a quincunx and a pair. Perhaps, in a larger domino set, e.g. "twelve" as a 3x4 rectangular arrangement may be perceived as subitizable numbers of rows and columns, combined with the learned knowledge of the product 3x4. But, by and large, I think we are outside the domain of subitizing here - not (merely) because the numbers of elements is too large, but because a different mental mechanism, much more dependent on training, is involved.--Nø (talk) 08:20, 10 May 2021 (UTC)

Five items can be subitised?
I'd like to contest that subitising only works to four. In my experience, it works up to five. I made my own game to test this, showing circles of different sizes and colours randomly scattered across the screen for 250 milliseconds. Each "round" shows between 1 and 10 circles at random, and the game lasts for 150 rounds. I played this game many times, and each time I got 100% accuracy between 1 and 5 items. Accuracy fell dramatically at 6 (and I was really just guessing if there were 7 or more items). Looking around in general, subitising definitely works to 5 for me. My friends played my game, and some of them also got 100% accuracy on 5 - although interestingly others saw the same dramatic accuracy fall at 5 and I saw at 6. I guess maybe subitising limit is down to the individual, so a maximum of four isn't a hard limit for all people. Minecraft69 (talk) 21:39, 11 August 2023 (UTC)


 * This is absolutely correct. In fact, the article does say, "By the age of seven that ability increases to 4–7 objects." I am personally not aware of research confirming that people can subitize six or seven randomly objects, but there is definitely research confirming that many elementary school students can subitize up to five. If there is somewhere in the article where it says people cannot subitize to five, that should be corrected.


 * I agree; the lead said
 * Sets larger than about four items cannot be subitized unless the items appear in a pattern with which the person is familiar
 * I've changed "about four" to "about for to five". Nø (talk) 18:05, 13 August 2023 (UTC)
 * However, "about four" could mean e.g. "3, 4 or 5", which, including young children, could be correct. "About four to five" is, arguably, an oddity. I suppose it could be taken to mean something like "4 or 5, or possibly 3 (or even less), or 6 (or even more)" - and that, I believe, is the correct amount of ambiguity in this statement.
 * Hvaing a source for such a statement would be good, though!
 * Perhaps, we could say "more than about 4, with significant individual variation"? Nø (talk) 08:26, 14 August 2023 (UTC)