Come and see you shovelers!
If you have a side herder at home:
(Image source: Weibo)
If you have a husky at home:
(Image source: Weibo)
Bian Mu's intelligence and husky's humble attitude can be regarded as recognized by everyone. However, some people who love Ha believe that huskies are not really stupid to learn instructions, but are eclectic, laissez-faire, and unwilling to be constrained by instructions.
Although there is no definite conclusion about the intelligence of huskies and border herders, scientists have also designed some methods for assessing the intelligence of animals.
How do I get the animals to answer the questions voluntarily?
First of all, we have to face a problem: animals can't listen, speak, read, write, and make scrolls, so how do we test their intelligence? To measure their intelligence, the scientists tailored "intelligence test rolls" to various animals. But how do you get animals to answer questions voluntarily?
To achieve this, scientists often need to train animals before a formal exam to develop the habit of answering questions. The most common way to train is to give food rewards.
For example, when pigeons peck at a picture, or when an orangutan touches a picture on a screen, they get a favorite food as a reward. At this stage, the animals have prizes as long as they do the problem, whether right or wrong. But after completing the pre-test training, the animals need to earn rewards based on "scores".
Chimpanzees can touch screen numbers sequentially after training
(Photo source: Kyoto University Primate Research Center)
So, what does the animal's "intelligence test" test test? Will there be a division of disciplines?
In fact, animal intelligence can be defined as the ability of an animal to change behavior, adapt to a new environment, or solve a new problem based on an individual's life experience. Intelligence encompasses a variety of cognitive abilities, and the "Animal Defined Volume" can be divided into "disciplines" according to the type of cognitive function.
Common cognitive skills subjects include memory, space-time perception, attention, reasoning, conceptual cognition, etc. Each subject has multiple questions to confirm each other, and here are a few representative subjects.
Module 1: Memory level test
The most abundant test subjects are the memory level test. The exam includes mazes, delayed response tests, and more. Among them, the maze is mainly used to test the spatial position memory ability of rats and mice, and is also often used to detect the effect of drugs on spatial memory.
There are many types of mazes, with different shapes and sizes, and the more commonly used ones are radiant arm mazes, Y mazes, T mazes, water mazes, Barnes mazes, etc. What these mazes have in common is that the critters explore within a fixed range of space and then find a man-made end point. The performance of the critter is determined by parameters such as the path during the exploration process and the time required to reach the end point.
Different maze principles are not the same, here are mice as an example to highlight several commonly used:
Topic 1 – T Maze / Y Maze
T maze
(Image source: Maze engineers)
Both the T maze and the Y maze consist of three channel arms. The maze explores with a gourmet temptation mouse hidden at the end of the arm of the passage. Before the experiment begins, the experimenter will seal the lower channel arm with a baffle and place it into the mouse. After removing the baffle, they follow the taste forward to find food.
Although the operation of the experiment is very simple and the negative impact on the animals is weak, the results of the experiment may be affected by the small animal's preference for a certain direction. For example, mice that are born with a preference for the left may be "indifferent" to food placed on the right maze arm.
Title 2 – Barnes Labyrinth
Barnes Maze
(Image source: Wikipedia: Barnes maze)
The Barnes Labyrinth primarily exploits the animal's darkening instincts. The nocturnal participants explored a huge disc, trying to find a "hidden corner" to hide from overhead light. There are many holes of the same size at the edge of the disc, but only one can lead to the cassette to avoid light.
The negative impact of this experiment on animals is also very weak, and it can more realistically and comprehensively examine the animals' memory levels. Because "candidates" face more choices in the test than in the three-channel maze – even blind guessing has a 50% chance of getting it right. But in this experiment, the participants are also prone to explore around and lack the motivation to complete the experiment. A few daring test takers will even run to the edge of the disc to experience the excitement of high-altitude overlooking.
Topic 3 – Water Maze
Water Labyrinth: Clear water on the left, visible below the circular floating platform
For the experiment on the right, the dye was added to the water
(Image source: Google Image)
The water maze experiment was conducted in a large "bathtub" with a circular floating platform in the bathtub. During the experiment, the water is stained. During training, the mice swim and explore, sometimes discovering submerged floats on their own. If it is not found for a long time, the experimenter will guide the participant to the floating platform.
There is a logo around the "bathtub", and the next time the experiment is performed, the mouse can find the floating table faster through the logo. But why do mice look for the float even if they can't see it? Mainly because the mice are naturally water-weary, while continuous swimming consumes energy quickly. In other words, if we are tired of swimming, we also want to find a place to rest as soon as possible.
Topic 4 – Delayed Response Test
In addition to various mazes, the discipline of memory also includes delayed response tests. It mainly detects the animal's short-term memory ability by the accuracy with which the animal delays the response to the signal. The subject animals receive a signal (sound, light). This signal corresponds to a condition for obtaining a reward.
For example, in an auditory decision-making experiment, the low-frequency sound corresponds to the water inlet on the left to drink, and the high frequency corresponds to the right. Mice are first trained for a single sound, and the two correspondences are learned separately through the monophonic training.
The low-frequency sound corresponds to water in the left nozzle, and the high-frequency corresponds to the right
(图片来源:Xin, Y., Zhong, L., Zhang, Y., Zhou, T., Pan, J., & Xu, N. L. (2019). ）
The process of formal experimentation is a bit like a CCTV "listening to songs and reading songs" variety show - "Opening the Door". The animals first hear a sound, pause for a while, and let them choose whether to be left or right. It's like the contestants recall the song after listening to it and choose which door to open. There is no reward for mice that choose the wrong direction or answer quickly. Delayed response testing has been widely used in many different species, including rodents, primates, birds.
Module 2: Primate Cognitive Skills Comprehensive Test
If the memory subject is a knowledge point variation test, then the primate cognitive ability comprehensive test (PCTB) includes the knowledge points of each subject. It comprehensively examines the comprehensive quality of animals, and the examination time is longer. PCTBs are used not only in primates, but also in birds such as parrots and other non-primate mammals such as mice and dogs.
PCTB questions cover both psychological and physical levels. At the physical level, there are three general directions of space, quantity, and causal reasoning. The psychological dimension includes social learning, communication, and the mind. Each general direction has 3-5 small directions, which can be described as all-encompassing. The advantages and limitations of this are equally obvious. The pctb has a variety of tests that corroborate each other. By correlating animal behavior with multiple cognitive criteria, scientists were able to gain a more complete picture of the animal's cognitive level.
For example, by analyzing the grain storage behavior of Eurasian grouse in winter and the correlation between their spatial memory ability, inhibition control ability, and classification learning ability measured by the PCTB, we may infer the relative importance of these abilities in grain storage behavior.
However, due to the cumbersome, time-consuming and laborious PCTB process, the requirements for animal numbers and experimental paradigm design are high, and it is only suitable for a few species. At the same time, because of the high requirements, the same set of PCTB experiments is rarely repeated, resulting in limited credibility.
Topic 1 – Use of Tools
In the different test experiments used in the PCTB, the use of tools was once considered a human-specific ability. However, the animals tested by the use of tools, in addition to our close relatives, non-human primates such as gorillas, there are elephants, many carnivorous mammals and crows and their close relatives.
A common test method is for researchers to give the animal a problem and provide the relevant tools to see if the animal will use the tool to "answer". To give a classic example: if you look at the picture below, the fruit elephant on the tree cannot reach, but a wooden box has been prepared next to it. The results of the experiment found that in order to reach the branches, the elephant would move the wooden box to the bottom of the tree to raise the forelimbs.
The elephants themselves "moved the stool" to reach the branches for a good meal
（图片来源：Foerder, P., Galloway, M., Barthel, T., Moore III, D. E., & Reiss, D. (2011).）
Another example of the use of the tool is the trap pipe test, which can detect both the animal's reasoning ability and problem-solving ability. Tests are mostly used in crows and primates. Just like the image below, the experiment uses a transparent pipeline. Transparent pipes are placed in the gorilla's cage, and the ends of the pipe connect the two sides of the cage.
When the orangutan is free to move inside the cage, it can reach into the sticks at both ends of the pipe to obtain the food in the middle. However, there is a square trap under one side of the pipe. Orangutans need to push food out of right to left with a stick or pull it back before it can be taken out of the pipe and eaten. If the food falls into the trap, the subject animal will have to be in a hurry.
If you push the food to the right with a stick, you will fall into the trap
(Image credit: Mulcahy, N. J., & Call, J. (2006).)
Similarly, there is a "cover drawer" experiment designed for crows. The main test is whether the crow can teach itself and master the use of the drawer. Scientists have tailored a small drawer for the crow. The side of the drawer replaces the handle with a "mouth". The crow needs to hold the ring on the side of the drawer with its mouth and pull it back in order to get food.
The crow will use its mouth to pull out a drawer to get food
（图片来源：Tebbich, S., Seed, A. M., Emery, N. J., & Clayton, N. S. (2007). ）
Topic 2 – Quantity Identification
In addition to the several tests introduced earlier, another one we are familiar with is quantity recognition. Although unfamiliar with Arabic numerals, many animals can identify and compare the number of them. Some intelligent animals, such as macaques, can also calculate simple additions and subtractions.
A parrot named Alex is famous for being able to count from 1 to 7 in English. Scientists speculate that it can recognize larger numbers, only to 7 due to the limitations of language expression. The quantity recognition test usually involves animals comparing the number of different elements in the picture, and also gives some different colored blocks like the following picture. (Parrots may be smarter than you think, poke here to get more interesting knowledge about parrots)
Alex the Parrot can count 7 objects in English
(Image source: Google Image)
Module 3: "Reading Comprehension" – Psychological Test
The previous introductions are all physical test questions, the test indicators are relatively objective, and the scientists' "grading papers" are also relatively good at scoring. In contrast, the psychological level of the test is more open to reading.
For example, social learning is usually a puzzle similar to the trap pipe experiment in the previous article. Before the test, the researchers demonstrate the answers to the puzzles and see if the subject animals learn from the demonstration. But sometimes it's hard for researchers to tell if they really learned in a demonstration or if they happened to succeed.
Another famous "subjective question" is the mirror test. The testing process itself is very simple: just make a mark on an invisible part of the animal's body that can be touched, such as a face, and observe how the animal reacts by looking in the mirror. The researchers then speculated whether the animals could recognize themselves based on their reactions. If the animal recognizes itself in the mirror, it is likely to touch the marker in the mirror.
When the mirror test was first invented 50 years ago, there were still relatively few animals tested. And only humans and some chimpanzees over 18 months old can pass, but in 2018, eurasian magpies also successfully passed the mirror test. Macaques, dogs, etc. cannot pass this test, and they will regard the mirror themselves as their kind or enemies. The inventor, Gallup, believed that the essence revealed by this test was self-awareness, and believed that this was a relatively rare high-level intelligence.
Dogs can't recognize themselves in the mirror
(Image source: gfycat.com)
However, the interpretation of the mirror test results is still controversial. On the one hand, the mirror test relies on vision, which is very unfair to animals that do not need vision in many life scenarios, and there is a suspicion of "discrimination" against some candidates. On the other hand, many animals have never seen a mirror, and they do not immediately understand the function of the mirror. Pigeons and macaques, for example, but after being trained to reflect images in mirrors, they can show what is called "self-awareness."
Elephants and "themselves" have a lot of fun
(Image source: gfycat.com)
However, there are also some psychological level test questions that have clear criteria for judging, such as gaze and follow the experiment. The experiment observed whether the animals would turn their heads from side to side and look in the corresponding direction; or by pointing to the cup experiment, the animals were tested to point to the cup containing the food as prompted by the researcher. But whether it's primates or crows, the results of these experiments hover around the passing line, which is almost the same as the correct rate of random guesses.
In the future, it may be more accurate
Tests that test the physical level of cognition in animals, such as the memory-related test highlighted above, have been widely used. In contrast, there are not many cognitive level tests at the psychological level, and there are often problems such as controversial or difficult to repeat results. An important reason is that animals can't speak, and it's hard for us to understand animal language. Animals can only be guessed based on their behavior, which is much harder than "You Draw Me Guess".
But with the continuous development of brain-computer interface technology and computer computing power, scientists in the field of neurology are also deepening their understanding of animal brains. I believe that in the near future, we can analyze the brain electrical activity of animals when doing various sets of "intelligence test papers" to understand the cognitive level of animals more carefully. Maybe at that time, the true face of Husky intelligence can be revealed by science.
Author Affilications:Chinese Academy of Sciences Brain Science and Intelligent Technology Excellence and Innovation
Copyright note: Unauthorized media reprinting and excerpting are strictly prohibited, and reprinting to platforms other than WeChat is strictly prohibited!
The article was first published in the Science Compound and represents only the author's views and does not represent the position of the Science Compound. Please contact the [email protected] for reproduction
The reproduced content represents the views of the author only
Does not represent the position of the Institute of Physics, Chinese Academy of Sciences
If you need to reprint, please contact the original public account
Source: Science Compound
EDIT: just_iu