Limits...
Trained Quantity Abilities in Horses (Equus caballus): A Preliminary Investigation.

Miletto Petrazzini ME - Behav Sci (Basel) (2014)

Bottom Line: Once believed to be a human prerogative, the capacity to discriminate between quantities now has also been reported in several vertebrates.The subject learned the discrimination in all conditions, showing the capacity to process numerical information.When presented with a higher numerical ratio (2 vs. 4, Test 2), the subject still discriminated between the quantities but its performance was statistically significant only in the non-controlled condition, suggesting that the subject used multiple cues in presence of a more difficult discrimination.

View Article: PubMed Central - PubMed

Affiliation: Department of General Psychology, University of Padova, Via Venezia 8, 35131 Padova, Italy; E-Mail: mariaelena.milettopetrazzini@unipd.it ; Tel.: +39-049-827-7424;

ABSTRACT
Once believed to be a human prerogative, the capacity to discriminate between quantities now has also been reported in several vertebrates. To date, only two studies investigated numerical abilities in horses (Equus caballus) but reported contrasting data. To assess whether horses can be trained to discriminate between quantities, I have set up a new experimental protocol using operant conditioning. One adult female was trained to discriminate between 1 and 4 (Test 1) in three different conditions: non-controlled continuous variables (numerical and continuous quantities that co-vary with number are simultaneously available), 50% controlled continuous variables (intermediate condition), and 100% controlled continuous variables (only numerical information available). The subject learned the discrimination in all conditions, showing the capacity to process numerical information. When presented with a higher numerical ratio (2 vs. 4, Test 2), the subject still discriminated between the quantities but its performance was statistically significant only in the non-controlled condition, suggesting that the subject used multiple cues in presence of a more difficult discrimination. On the whole, the results here reported encourage the use of this experimental protocol as a valid tool to investigate the capacity to process numerical and continuous quantities in horses in future research.

No MeSH data available.


Related in: MedlinePlus

Accuracy (proportion of correct choices) is plotted against the type of stimuli (non-controlled, 50% controlled, 100% controlled stimuli) separately for each Test. Asterisks (*) denote a significant departure from chance level. Horizontal dashed line indicates chance level.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4219264&req=5

behavsci-04-00213-f002: Accuracy (proportion of correct choices) is plotted against the type of stimuli (non-controlled, 50% controlled, 100% controlled stimuli) separately for each Test. Asterisks (*) denote a significant departure from chance level. Horizontal dashed line indicates chance level.

Mentions: Shanty reached the learning criterion after 132 trials. The frequency of correct choices was statistically significant (chi-square, χ(1) = 29.121, p < 0.001, Figure 2). In particular, Shanty proved able to discriminate the two quantities in all of the conditions (non-controlled stimuli: χ(1) = 13.091, p < 0.001; 50% controlled stimuli χ(1) = 9.091, p = 0.003; 100% controlled stimuli χ(1) = 7.364, p = 0.007). No left-right bias was observed (proportion of left choices, 0.59; proportion of right choices, 0.41, χ(1) = 2.979, p = 0.084).


Trained Quantity Abilities in Horses (Equus caballus): A Preliminary Investigation.

Miletto Petrazzini ME - Behav Sci (Basel) (2014)

Accuracy (proportion of correct choices) is plotted against the type of stimuli (non-controlled, 50% controlled, 100% controlled stimuli) separately for each Test. Asterisks (*) denote a significant departure from chance level. Horizontal dashed line indicates chance level.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4219264&req=5

behavsci-04-00213-f002: Accuracy (proportion of correct choices) is plotted against the type of stimuli (non-controlled, 50% controlled, 100% controlled stimuli) separately for each Test. Asterisks (*) denote a significant departure from chance level. Horizontal dashed line indicates chance level.
Mentions: Shanty reached the learning criterion after 132 trials. The frequency of correct choices was statistically significant (chi-square, χ(1) = 29.121, p < 0.001, Figure 2). In particular, Shanty proved able to discriminate the two quantities in all of the conditions (non-controlled stimuli: χ(1) = 13.091, p < 0.001; 50% controlled stimuli χ(1) = 9.091, p = 0.003; 100% controlled stimuli χ(1) = 7.364, p = 0.007). No left-right bias was observed (proportion of left choices, 0.59; proportion of right choices, 0.41, χ(1) = 2.979, p = 0.084).

Bottom Line: Once believed to be a human prerogative, the capacity to discriminate between quantities now has also been reported in several vertebrates.The subject learned the discrimination in all conditions, showing the capacity to process numerical information.When presented with a higher numerical ratio (2 vs. 4, Test 2), the subject still discriminated between the quantities but its performance was statistically significant only in the non-controlled condition, suggesting that the subject used multiple cues in presence of a more difficult discrimination.

View Article: PubMed Central - PubMed

Affiliation: Department of General Psychology, University of Padova, Via Venezia 8, 35131 Padova, Italy; E-Mail: mariaelena.milettopetrazzini@unipd.it ; Tel.: +39-049-827-7424;

ABSTRACT
Once believed to be a human prerogative, the capacity to discriminate between quantities now has also been reported in several vertebrates. To date, only two studies investigated numerical abilities in horses (Equus caballus) but reported contrasting data. To assess whether horses can be trained to discriminate between quantities, I have set up a new experimental protocol using operant conditioning. One adult female was trained to discriminate between 1 and 4 (Test 1) in three different conditions: non-controlled continuous variables (numerical and continuous quantities that co-vary with number are simultaneously available), 50% controlled continuous variables (intermediate condition), and 100% controlled continuous variables (only numerical information available). The subject learned the discrimination in all conditions, showing the capacity to process numerical information. When presented with a higher numerical ratio (2 vs. 4, Test 2), the subject still discriminated between the quantities but its performance was statistically significant only in the non-controlled condition, suggesting that the subject used multiple cues in presence of a more difficult discrimination. On the whole, the results here reported encourage the use of this experimental protocol as a valid tool to investigate the capacity to process numerical and continuous quantities in horses in future research.

No MeSH data available.


Related in: MedlinePlus