Moderated by Stephen Muggleton.

Zhanna Reznikova and Boris Ryabko

How to study ants. Numerical competence using their own communicative means and applying ideas of information theory

The article mentioned above has been submitted to the Electronic Transactions on Artificial Intelligence, and the present page contains the review discussion. Click here for more explanations and for the webpage of theauthors: Zhanna Reznikova and Boris Ryabko.

Overview of interactions

N:o Question Answer(s) Continued discussion
1 15.10  Anonymous Referee 1
2 15.10  Anonymous Referee 2

Q1. Anonymous Referee 1 (15.10):

The paper should be accepted subject to major revisions.

The title could with advantage be shortened to "A Study of Ants' Numerical Competence"

Substantive required revisions include a careful check by the authors that all the points in the following Referee's Specimen Summary have been clearly stated (as part of the text, not just in captions of Tables or Figures) in the re-submitted paper (in the text, not just in captions of Tables or Figures).

Referee's specimen summary of main points

The paper is in two parts.

The first part describes an elementary maze in the form of a main street with 30 or more side-streets, at the end of one of which a food reward is hidden. Some of these "main streets", referred to in the paper as trunks, were vertical, some horizontal and some were circular.

Experiments using this set-up are reported in which the durations of interactions between scout ants and their foraging teams were measured.

The magnitude of the integer that has to be transmitted in these interactions to enable the foragers to find their way unaided to a hidden food source is related to these durations. Results of numerous experiments have lent strong support to the hypothesis that integers are transmitted using a unary number system. Results of a selected sample experiment are tabulated and graphed.

The second part describes a more complex set up, in the form of m x n grids (city-block style). Here too the results are strongly supportive of the hypothesis that the scouts use the interaction sessions to convey symbolically the whereabouts of the distant food. Possibly the ant messages express the whereabouts as a pair of integers corresponding to the food's spatial co-ordinates. Further analyses are proceeding to test this hypothesis on the assumption that each member of the co-ordinate pair is expressed in unary, differing only from the scheme described in Part 1 in requiring a pair of integers to be transmitted rather than just one.

Broad proposals for revision

Diagrammatic pictures are necessary for understanding the "trunk", "circle" and "lattice" mazes.

The second Part of the paper, concerning the lattice mazes, should drop the "globe" variant, apart from mentioning that it was used in separate experiments not reported here. Inclusion in this paper would complicate the discussion, and distract from the more clear-cut topic of numerical co-ordinates.

All detail from this second part should be dropped, including the Tabulation. The work with grid mazes should only briefly be described, as work in progress.

This particular investigation cannot be regarded as completed until the authors have found the time to prepare a summarizing Figure, in which message-durations are plotted using the sum of the two co-ordinates to label the horizontal axis. This style of presentation would then be comparable to that which needs to be done in the first Part to help visualize the content of Table 1 (see below).

Detailed proposals for revision

Part 1

Table 1 should be supplemented by a Figure in which the Table's contents are re-expressed in graphical form. Durations of messages should be plotted against against magnitudes of integers.

The opening passage has no Subtitle. If it is intended as an Abstract, then it needs shortening, for example as follows

Specimen condensation of Abstract

Our experiments provided a situation in which insects had to transmit information quantitatively known to the experimentalist, in order to obtain food. One may measure the complexity of tasks they solved in order to pass definite pieces of information from scout ants to foragers.

Our previous experiments, based on ideas of Information Theory, had - - shown that ants [name the species] are able to memorize and transmit - messages concerning sequences of turns towards a hidden trough of syrup and to use the simplest regularities to compress the information.

In the experiments here described counting and number-related skills are described. We selected aphid-farming red-wood ants of Formica polyctena to transmit information (1) on the ordinal number of a given side-branch in a linear maze and (2) on the numerical co-ordinates in a grid-type maze of a node at which syrup had been hidden. [Note suggested use of hyphen to distinguish "red-wood ants" from "red wood-ants".]

One of the experimental set-ups consisted of a "tree trunk" with branches that ended in empty troughs, except for one which was filled with syrup. Another set-up consisted of a lattice or grid, in which location of a goal-node relative to a start-node can be specified as a pair of Cartesian coordinates. The foragers of F. polyctena were separated into teams of 5-8 individuals, each with one scout. All laboratory ants were marked with coloured labels. To start the experiment, an ant scout was placed at a randomly chosen trough containing food and then allowed to return to the nest on its - own. The duration of the ensuing contact between foragers and the scout was measured. Then we removed the scout and the foragers had to search for the food by themselves. The experiments were so devised as to eliminate all possible ways that may help to find food, except for - distant homing. The consistently high success of the foragers in finding the food suggests that scouts are able to count within a range of several tens, and to transmit this information to their foragers. Analysis of time durations of ants' contacts enable us to hypothesise how they use numbers and numerical coordinates in their communication.

The last two sentences of the original are dropped from the above condensed version, being of a purely speculative nature.

Since this paper covers numerical competence, needing no significant use of information theory for its description, reference to information can be dropped from the paper's title and from the body of the paper. It is only necessary to retain the following passage after deleting the first sentence - which is redundant - and making minor textual clarifications, as below.

"This allows us to estimate intellectual potentials by observing the communicative process. Other experiments based on ideas of Information Theory have shown that ants probably have an even more intricate form of communication than the honeybee. As reported here, we also succeeded in studying numerical properties of insect cognitive capacities, namely their ability to perform limited counting and to memorize simple regularities, thus compressing the information available (Reznikova, Ryabko, 1993, 1994, 2000; Ryabko, Reznikova, 1996)."

Later, the passage "Then, 'incapable' scouts which either could not memorize well the way to the trough or could not mobilize their groups were revealed and discarded from further experiments." should read "... discarded before the main experiments were started."

The paper's last passage, beginning "In conclusion, we believe that ... " should be dropped. It offers no scientific facts or arguments in support of the stated belief. Therefore its inclusion will only add to the scientific reader's work without adding to his information.

Q2. Anonymous Referee 2 (15.10):

Recommendation: The paper can be accepted for ETAI after minor revisions. No further refereeing round is needed.

This is a facinating paper. It describes well designed and executed experiments to apply information theory to make inferences about the methods of communication used by ants. The authors suggest that similar methods may be used on other animals.

The main thing that needs to be done to improve the paper is to make corrections to the grammar. The English is generally clear and understandable, but there are frequent and consistent mistakes, particularly with the omission of articles.


Background: Review Protocol Pages and the ETAI

This Review Protocol Page (RPP) is a part of the webpage structure for the Electronic Transactions on Artificial Intelligence, or ETAI. The ETAI is an electronic journal that uses the Internet medium not merely for distributing the articles, but also for a novel, two-stage review procedure. The first review phase is open and allows the peer community to ask questions to the author and to create a discussion about the contribution. The second phase - called refereeing in the ETAI - is like conventional journal refereeing except that the major part of the required feedback is supposed to have occurred already in the first, review phase.

The referees make a recommendation whether the article is to be accepted or declined, as usual. The article and the discussion remain on-line regardless of whether the article was accepted or not. Additional questions and discussion after the acceptance decision are welcomed.

The Review Protocol Page is used as a working structure for the entire reviewing process. During the first (review) phase it accumulates the successive debate contributions. If the referees make specific comments about the article in the refereeing phase, then those comments are posted on the RPP as well, but without indicating the identity of the referee. (In many cases the referees may return simply an " accept" or " decline" recommendation, namely if sufficient feedback has been obtained already in the review phase).