1. Five links and three elements in the research.
There are five important links in the research work, including: raising questions, designing schemes, conducting experiments, sorting out data and writing papers. These five links must be strictly required in order to produce high-level research results. In these five links, I think asking questions and designing schemes are the most creative.
We should also pay attention to three elements in doing research: independence, systematicness and creativity. Independence means that the main part of the thesis is completed by graduate students themselves, and the guidance of teachers and the help of classmates are necessary, but they only play an auxiliary role. This includes independently proposing experimental design, independently operating experiments, independently processing data, and independently writing papers. Because of different knowledge backgrounds, individual graduate students can't design independently, can't process data, and have to ask others for help in everything, which is not acceptable. You must learn to do these things independently to become a qualified graduate student. Systematization means that the paper should systematically answer (or solve) one or several important academic questions from different sides, the structure of the paper should be rigorous, the workload should be heavy or heavy, and hard work should be paid; Creativity means that the paper obtains new results and conclusions and solves important problems that have not been solved by predecessors, which is meaningful and even important for the development of the discipline. In these aspects, the requirements for doctoral dissertations should be higher than those for master dissertations, and there should be differences in independence, systematicness and innovation.
2. Read the literature and write a literature review.
Reading literature is an important way to accumulate knowledge and an important basis for finding and asking questions. For graduate students, this is a basic training. From my own experience, those students who learn well are usually students who love reading literature, read a lot of literature and are very familiar with literature. They read more documents than I do, and they understand many things better than I do.
How to read literature? First of all, we should pay attention to the choice of literature, and read a certain number of high-end publications, which are of high level and cutting-edge research, and will be more inspiring to ourselves. I advocate reading high-end journals 100 papers from this perspective. We should also start from the recent literature, from near to far. Start with the new document, and find several (usually 4-6) related documents from the documents listed later for secondary reading. Then, based on these documents, find 3-5 more articles from the literature behind, and start to pursue them, so that you can read 20-30 related articles. In this way, it is easy to understand the origin and development of a problem, so as to grasp the problem systematically and help to remember these documents. We should also combine intensive reading with skimming. Good articles and high-level articles must be read intensively and carefully, including many details. The success of many studies is determined by these details.
When reading literature, we should use our brains, pay attention to analysis, "read articles analytically and think critically". I advocate reading in and reading out. By reading in, I mean really understanding the contents of the document, not being satisfied with a little knowledge. The so-called reading refers to transcending literary thoughts and forming one's own analysis and judgment. Many students read the literature and think that what others say is right and good. Without opinions and ideas, they have never read it. Reading documents should also be "consistent with knowledge", that is, finding the internal relations of many documents and synthesizing and summarizing them in a very concise and clear way. There are many models in cognitive psychology, which are put forward according to the principle of "from cognition to consistency, from consistency". Many high-level research results can often be expressed in a very concise way. If we can read the literature, we will reach the highest level.
Reading literature should also pay attention to writing literature review, including notes, experiences and articles. Writing literature review is the basic skill of doing research. Keep writing and move forward step by step. This process is a process of accumulating knowledge and forming personal "ideological system", so it is also an important way to choose scientific problems and determine the research direction. Now some students don't pay attention to writing literature review, and the teacher takes out a PPT file when checking, which is not acceptable. PPT file is a way to express knowledge, and replacing literature review with PPT file will affect our in-depth understanding of literature. I ask students to write a literature review, which PPT files can never replace.
3. Be good at putting forward and practicing scientific problems.
Putting forward valuable scientific questions is the premise of doing research. Many graduate students will stop asking questions after they have determined a topic to study. There are several situations here: 1) The meaning and value of the question is not great, and the question raised can be done or not; 2) The problem is vague and I don't know where the core of the problem lies; 3) Lack of internal logic and unclear relationship between several problems; 4) The problem is too big or too small. For example, we asked a student to study the influence of computer experience on human brain processing and word processing functions by comparing different input methods of computers. In the first few drafts of the paper, she asked "What is the writing center"? This problem is obviously practical and does not fully reflect the important scientific significance of the research in the title. After many discussions and revisions, the problem is defined as "the influence of computer input mode on the language function of human brain". As we all know, the use of computers has changed many aspects of human life, including the function and structure of the human brain. If our research can explore the influence of computer use on human brain function, it will be a very meaningful thing, which is more scientific than finding a "writing center". Not long ago, I participated in a doctoral thesis opening report. She hopes to study the influence of working memory capacity on second language syntactic processing, but her topic is the influence of working memory on second language syntactic processing. Besides capacity, working memory also contains some other functions. Her original topic was obviously too big.
Questioning should run through all aspects of research, including data processing and article writing. A graduate student studied the influence of mother tongue experience on the second language in his graduation thesis. Later, when rearranging the materials and writing articles, she raised a new question: the role of the left caudate nucleus in bilingual control, which made her rethink a new direction, re-read and reorganize the documents, re-process the materials and revise the articles many times. Finally, she wrote a high-quality article and published it in internationally renowned academic journals. In this process, the problem has played a guiding role, making the research more and more in-depth.
4. Ingeniously design experiments
I divide the research into three categories: one is confirmatory research, which verifies some results and conclusions of predecessors through my own research. This kind of research is valuable when we are beginners. The second kind of research is local innovation research. Most of the research we are doing now falls into this category. Some of these studies are based on one or several articles, asking new questions or developing new paradigms, which are based on predecessors and higher than predecessors, and get some new results and conclusions. The third category is original innovation research. This kind of research should stand at a new theoretical height, ask new questions, and get important research results that have not been obtained by predecessors. The articles we read in top academic journals such as Nature, Science, Neuron and PNAS are often original and innovative.
Experimental design is the most creative and important link in research work. Design experiments should focus on the overall situation and start from small things. Focusing on the overall situation means that research should have macro thinking, think bigger about the problem and tell bigger stories; At the same time, the starting point of research should be small and the problems should be concentrated. Only in this way can we do the work well.
How to design the experiment? Many students usually make two mistakes when designing experiments. One is that they are large and complete, and there are too many variables set in an experiment. I hope that many problems can be solved simultaneously through an experiment. Because there are too many variables and there are multiple interactions between variables, the results are very complicated and difficult to explain. Another problem is that in a series of experiments, two adjacent experiments change not one condition, but more than two conditions, so it is not clear which factor leads to the result. The following is a letter I gave to my students, mainly aiming at the first question: "The psychological phenomenon of people is very complicated, and how to study it scientifically is a very difficult problem. Many beginners like to design experiments very complicated, thinking that only complexity can detect the internal mechanism of psychological phenomena; And a mature researcher will design the experiment very simple or relatively simple, and answer more complicated questions through a series of seemingly simple experiments. As a result, the latter succeeded, while the former failed and was trapped in the mud pit dug by himself. Therefore, when designing experiments, the more complicated the better. We should try our best to break down complex problems into relatively simple ones and then solve them one by one. Only one problem can be solved in each experiment, and more complex problems can be solved by accumulating many experiments. This is the basic requirement and secret of designing experiments. If this idea is not solved, it will be difficult to enter the research field of psychology, let alone do basic research. "
5. Plans and articles should be revised repeatedly.
People's understanding of things often needs to go through a process from shallow to deep, from the outside to the inside. The same is true of doing research. It is impossible to plan, design and write a good article at one time, and it is necessary to try and revise it again and again. Anyone who has published an article knows that an article has to be revised many times from the first draft to publication, ranging from 5 or 6 times to 10 times. Some of our articles published in international journals have to be revised many times before submission, and they have to be revised repeatedly according to the opinions of reviewers after submission, and finally they can be published. In this process, we must have a little persistence and don't give up easily. Be good at affirming yourself and dare to deny yourself. Being good at affirming yourself means having confidence in your own research, sticking to what you should do, and not giving up your original ideas easily after hearing other people's opinions. In reviewing papers, some reviewers' opinions are very sharp and sometimes harsh. If they are not good at dealing with their problems, the articles may be rejected. Therefore, it is not easy to be good at affirming yourself. At the same time, it is even harder to dare to deny yourself. It is often difficult for a person to find his own problems. Sometimes he knows that there are problems, but he still has to protect his weaknesses. Therefore, you can't deny your own plan and your first draft. In research work, people are often easily influenced by "inertia" and are unwilling to change their plans and write literature reviews. At this time, it is particularly important to break the "inertia" and think from another angle.
Here I recommend an article I like very much: The Hill in the West of Kentan by Liu Zongyuan, a famous scholar in the Tang Dynasty. This article is not an article, but it is very enlightening for us to write articles. The full text is as follows:
After eight days in the West Mountain, I found 200 steps northwest of the mountain pass, and I arrived at Lake Copomb. Twenty-five steps to the west of Cobham Pool is the dam deep in the current. There is a hill on the top of the dam with bamboo and trees growing on it. The stones on the mountain stand out, and then the earth breaks through the ground, competing for competition, and their number is almost unknown. People who are tired of each other are like cows and horses drinking streams; Those who rush to the top of the column are like bears climbing on the mountain.
The hill is very small, less than an acre, and can be put into a cage for your own. Ask its owner, saying: the abandoned land of Tang Dynasty. Ask the price and say: only 400. I like it. I'll take it. I have been there with Li Shenyuan and Yuan Keji, and they are all very happy and have an unexpected harvest. Then take turns to pick up tools, shovel weeds, cut down trees and light a fire to burn them. The tree is beautiful, the bamboo is beautiful, and the stone is wonderful. Seen from above, the mountains are high, the clouds are floating, the streams are flowing, and birds and animals are swimming with them. Give me a call and give me a clever trick to help you at the foot of this mountain. Lying on the pillow, you will understand the form and sight, sound and ears, the leisurely and empty man and god, but the deep and quiet man and heart. Two scenic spots came in less than ten days, and even ancient mountain lovers may not have been to this place.
Hey! With the victory, Feng, pickaxe, Kun and Du are brought to you, and expensive tourists compete for buyers, which is getting harder and harder to get. Now abandoned by the state, the farmer's fish father is poor, the price is 400, and he can't sell it for years. However, I am only happy with God and self-denial, and the result is not good. Write this article on a stone and congratulate the hill.
Our experimental scheme, the experimental data obtained and the first draft of the article written are sometimes like the hills in the west of Kemtan. At first glance, they are overgrown with weeds, disorderly and worthless. Only after some processing and pruning, "shovel dirty grass, cut evil wood and burn" will the beauty of the article be highlighted, making it "beautiful, beautiful and strange." Of course, the article must be really beautiful, and fire and logging should be used properly, otherwise, after logging and "fire", only ashes and ruins will remain. We should let students try their best to discover the beauty of schemes and articles, and also help students discover the beauty of their own works.
6. Pay attention to basic skills training
Many "basic skills" problems are often found in thesis writing. These questions sometimes surprise me. I didn't expect graduate students in our State Key Laboratory to make such low-level mistakes. Problems include: 1) The purpose of the experiment is not clear. 2) The research logic is unclear, the research scheme and experimental design are not demonstrated, and statistical treatment is not carried out according to the experimental design. 3) The table is not standardized, and there is no description of the header and figure. The chart is inconsistent with the text. 4) No experimental materials are provided in the appendix. 5) Can't write discussion. The discussion only repeated the results of the experiment, without in-depth explanation, simple and superficial. 6) You can't write a conclusion, and use the result description instead of the conclusion. 7) The documents are disordered and irregular, and the documents cited in this paper can't be found behind.
In short, for a person who is interested in research, doing research is his life's work. We should combine lofty ideals with down-to-earth spirit and dare to be the first. People's abilities vary widely, and people's hobbies vary widely. But as long as you have a clear goal, work hard and seize the opportunity, you will certainly achieve something in your career and realize your ideals and values in life.