Teaching your dog tricksPart of · Teaching your dog tricks

Teach your dog the names of objects

8 min read

Teaching your dog the name of an object comes down to one recipe: name the object just before each retrieve, dozens of times over several days, until the word and the thing become one. Then you mark the right choice and ignore the mistake, never a "no".

Why it works (and what we can promise)

For your dog, linking the sound "lead" to an object is a form of referential memory: the word becomes the stand-in for the thing. It is one of the most demanding pieces of learning in the whole repertoire, and a big mental expense: a few minutes of discrimination tire him out more than a long run.

Before you begin

This trick is an advanced level. Three basics need to be solid already.

0 / 3

The path, step by step

One object and one name at a time, learned on its own until solid, then mixed in. Each new object starts again from step 1. Count on several days per level, in short sessions.

Step 1 / 5

Un objet dédié et un nom dédié, faciles à distinguer.
- Take an object that looks very different from everything else (shape, size, texture), sized to your dog's mouth, with no small detachable parts.
- Give it a short, sonorous name, distinct from the cues he already knows.
Move on when: Ton chien rapporte déjà cet objet de façon fiable.
Le mot et la chose fusionnent.
- Present the object, say the name, let him orient towards it, take it, bring it back.
- Mark the orientation, then the pick-up, and reward.
- Repeat many times, spread over several days: the sound must "stick" to the object before any test.
Move on when: Au seul énoncé du nom, ton chien s'oriente franchement vers l'objet, plusieurs séances de suite.
Choisir le bon objet, pas le plus proche.
- Place beside it an object that has never been retrieved, never named, neutral and with no play value.
- Ask for the learnt name: mark the right choice, ignore the mistake (no "no", you simply try again, made easier).
- Change the position of the two objects on every attempt (left/right, near/far) to kill the spatial cue.
Move on when: Réussite franche et reproductible sur des essais où tu as varié les positions (de l'ordre de 5 fois sur 6).
Discriminer deux noms, sans deviner.
- Teach the second object exactly as in steps 1 to 3, in isolation.
- Once it is solid, present the two known objects together and ask for one then the other in a random order, with varied positions.
Move on when: Choix correct entre les deux, ordre tiré au hasard, avant d'ajouter un troisième.
Un vocabulaire qui se construit sans se brouiller.
- Add just one name at a time: the repertoire grows slowly.
- Start with objects that are easy to tell apart, then bring their appearances a little closer together (two toys of the same shape, different colours): this is where discrimination becomes fine.
Move on when: Chaque nouveau nom atteint le seuil de l'étape précédente avant le suivant.

The vocabulary is under way

The number of repetitions and the "5 out of 6" benchmark are practical guardrails, not measured constants. The solid principle lies elsewhere: a name is learnt through repeated association, one criterion at a time, at the dog's own pace and without coercion (AVSAB 2021), and the dog remembers better in short, spaced-out sessions than in long, daily ones (Demant et al. 2011).

The traps that sabotage everything

Saying "no" over a wrong choice: you punish the attempt, your dog starts to hesitate and ends up no longer committing. We mark the right one, we let the wrong one go.
Guiding without realising it: staring at the right object, turning your torso, taking half a step towards it. Your dog is reading you, not the name (this is the Clever Hans effect, Pfungst 1907). Stay neutral.
Always leaving the object in the same place: your dog learns the position, not the word. Mix up the location from the distractor stage onwards.
Introducing the second name too soon: before a first name is genuinely solid, the two labels get muddled.
Starting with objects that look alike: fine discrimination comes at the end, not at the start.
Promising "the dog who knows 100 words": you set up counterproductive pressure and you miss out on the real benefit.

Objects with no small detachable or swallowable parts, sized to your dog's mouth, nothing breakable, sharp or toxic (no batteries, pens or small rigid bits of plastic). To begin with, use objects dedicated to the trick, and a non-slip floor if your dog sets off at a run.

Kaminski, Call & Fischer — Word learning in a domestic dog: evidence for fast mapping (Rico), Science (2004)
Pilley & Reid — Border collie comprehends object names as verbal referents (Chaser, 1,022 names), Behavioural Processes (2011)
Fugazza et al. — Rare acquisition of object names in ordinary dogs (2021)
Dror, Miklósi, Sommese, Temesi & Fugazza — Rapid acquisition and lasting memory of object names in "Gifted Word Learner" dogs, Royal Society Open Science (2021)
Pfungst — Clever Hans: the reading of involuntary bodily cues (1907)
Demant et al. — Short, spaced-out sessions: better acquisition (2011)
AVSAB — Position statement on humane dog training (2021)

To go further

guideHold and carry an objectThe prerequisite: holding and giving back without chewing.Read the guide guideGo to Mum, take this to DadThe same idea, but for family members.Read the guide guideThe dog who "counts" and the Clever Hans effectWhy testing without being able to guide is the golden rule.Read the guide

Frequently asked questions

How many words can a dog learn?

Most dogs learn a handful of object names, and that is already a fine achievement. Vocabularies of hundreds of words (Chaser, 1,022 names, Pilley & Reid 2011) are the exception, often exceptionally gifted border collies. Aim for two or three useful objects rather than a record.

How can I teach my dog to fetch a specific object?

Name the object just before each retrieve, dozens of times over several days, until the word "sticks" to the thing. Then add a neutral object beside it, ask for the learnt name, mark the right choice with a "yes!" and ignore the mistake. Always vary the position of the objects.

How can I teach my dog to go and fetch his lead?

It is the perfect use case: the lead is unique and looks different from everything else. Once he retrieves it reliably, say "lead" just before each retrieve, mark, reward, and repeat over several small sessions. Soon "go and fetch your lead" will be enough.

My dog picks the wrong object, what should I do?

Above all, no "no": telling him off for the mistake teaches him to stop daring to try. You mark the right choice and let the wrong one go, while making things simpler (objects further apart, more different). If he gets it wrong often, it means you went too fast: go back to a single, well-solid object.

From what age can a dog learn the names of objects?

It is an advanced trick that calls for a retrieve and a "hold / give" that are already reliable, so rather a dog who is a little more mature. But there is no upper age limit: a dog learns throughout his life, and this mental work suits an older dog very well, in short sessions.

Does my dog really know the name, or is he reading me?

An excellent question, and the only way to find out is to test without being able to help him: objects in varied positions, and ideally someone who asks for the name without seeing where the right object is. Without that, you are often measuring your own involuntary body language (the Clever Hans effect), not his knowledge of the word.

Loading your progress…