In the early 1940's, the first practical and successful electronic computer was developed. A great variety of computing systems have since been devised, including electronic "digital computers" in which electronic circuits are utilized to perform the basic computing operations. Electronic digital computers typically operate in accordance with a selected set of arithmetic logic functions which may be performed extremely rapidly and with a very high degree of accuracy. Accordingly, in general, electronic digital computers are capable of highly complex and quite rapid mathematical and logical operations. More recently, computers have been developed in which a very large number of individual computing elements, or "n.times.m" cells, are arranged in a memory matrix. The n.times.m matrix is conceptually a rectangular array comprising a total of n rows by m columns. The matrix is utilized to represent a predetermined number of individual, unique binary numbers, the matrix itself containing no additional data. Each unique number is represented by a separate row of cells. Further, to make it easier to identify each unique number and, hence, each row, at least one column is assigned a preassigned binary number, termed a "tag," which identifies the row. A plurality of such n.times.m matrices can be assembled to form a very large array or "super-matrix." A super-matrix comprising 16.times.16, for example, could contain 4,096 rows and 4,096 columns. Each row, and each column, contain n cells, totaling 16.times.16 cells or 256 cells per matrix. A matrix of the size 16.times.16 could therefore be considered to be a very large "array" of 256 cells. The contents of a given cell in the matrix are determined, of course, by the sum of the contents of the other 255 cells which share that row and column. The sum of the contents of the cells in any row or column is referred to as the "index" of that row or column. This number can take on any one of a total of 2.sup.m (m.sup.2) different values. The "index" is the most significant bit of each word in the computer. These words are typically 32 bits in length and can, in a practical sense, contain only two values: zero or one. In some binary computing systems, one or two more bits (one or two "weights") may be utilized to give higher order accuracy to a weight. All of the words in the computer are, therefore, 2.sup.m (m.sup.2) times the length of the most significant bits in a given word. For example, assuming that the most significant bit in a word can represent only two values, the word "0101101" can represent four different numbers. In the decimal number system, each different number can be written as a decimal number in a different order (one, five, zero, four, three, six, two, seven, nine, etc.). However, in the binary number system, there is no equivalent way of writing each different number. All numbers in the binary number system are simply 2.sup.m (m.sup.2) different numbers, each represented by a unique m.sup.2 bit word. As noted previously, to make it easier to identify each unique number, at least one column is designated as having a tag which identifies the row. If there is a total of k unique numbers which are being represented by the matrix, the total number of columns designated as "tags" is k. Thus, in a matrix in which each row contains n cells, there will be n.times.k unique numbers. It is to be understood that the total number of possible rows for the above-described n.times.m matrix can be determined by the mathematical formula, n.times.m=2.sup.m.sup.m. Each row, and therefore each unique number, therefore has a unique binary number. Thus, if there are n possible unique binary numbers and each of these numbers is represented by m bits, there are n.times.m possible different binary numbers, or n.times.m.times.m=2.sup.n.sup.m different numbers, all having different binary representations. While an n.times.m matrix can represent up to 2.sup.n.sup.m unique numbers, there may be situations in which it is desirable to utilize such a matrix to represent a lesser number of unique numbers. For example, if only 16 numbers are to be represented, the numbers can be arranged in an n.times.m matrix in which the total number of columns, k, is sixteen. If there are sixteen numbers, then the total number of rows (and therefore the number of unique binary numbers) is 2.sup.m.sup.16, and therefore 2.sup.16.times.2.sup.16=2.sup.32 unique numbers. As mentioned above, a matrix of the size 16.times.16, for example, can be considered to be a very large "array" of 256 cells. In fact, any matrix of 16.times.16 or larger can be thought of as an array of 256 cells and the words or "bits" used in such a computer can be thought of as words of 256 cells. That is, in a 16.times.16 matrix, the computer may have words of 16.times.16 cells (256 cells), or even 17.times.17 cells (289 cells), or even 19.times.19 cells (719 cells), etc. When using such a word, and a matrix of 16.times.16, for example, then the contents of each cell in the matrix will depend only upon the number in the word which is located in the row and the column in which the given cell is located. For example, to represent the array of 256.times.256 cells (or words of 256 cells) in a matrix of 16.times.16, the bits representing the numbers in the matrix can be divided into four words each having 16 bits. Each word of 16 bits is used to represent the 256.times.256 matrix by providing the computer with an index word for each row and column, where the index word is the most significant bit of each word. That is, the entire computer is represented by the following matrix: ______________________________________ 16 bits in word 1 for row index 2 bits in word 1 for column index 16 bits in word 1 ______________________________________ In this example, each bit in a word of 16 bits represents a position in the array in which a zero or one exists. For example, the bit in the first column in the word of 16 bits may be designated as representing row 2, column 2. Thus, the bits in this column can only be zero or one (i.e., 00 or 11) and only in that position in the array. The row index and column index words in the matrix can be any value (e.g., 1, 2, 3, etc.) but typically they are values between 0 and 16, inclusive. Each cell in the matrix can be represented by the combination of a row index word and a column index word. For example, the cell in row 2, column 2, can be represented by the 16 bit word: "110011001100110010." Thus, in the 16.times.16 matrix, there are 256.times.256 cells in the matrix, and the contents of each cell depend only upon the row index and column index words of the computer. The present invention is concerned with matrix display devices in which each cell in the matrix contains a memory cell in which an electric charge can be stored. This memory cell may, for example, be a "memory transistor" of the type disclosed in commonly-owned U.S. Pat. No. 4,255,686. In any case, a cell must be capable of storing at least one "weight" or "voltage level" in the binary number system. It is, however, necessary for reasons of the complexity of such a matrix to be able to read and to write into each cell of the matrix. That is, if a user of the computer can write into each cell of the matrix, and has full knowledge of what each cell contains, then the user can manipulate the matrix to store his or her own information in the computer. For example, the user can provide a negative charge to a cell in the matrix in order to represent a zero, and the user can provide a positive charge to a cell in order to represent a one. In addition, a computer is particularly efficient if, for example, each column in the matrix has an associated memory cell which can be written to when the user desires to write into the column. This allows a user to store information which does not correspond to a row but rather to a column of cells in the matrix. This can be accomplished, for example, by providing for a write enable wire which can be coupled to a particular column in the matrix. Then, when a user desires to write information into a particular column, the write enable wire is activated. Such a column is said to be "active." If the write enable wire is not activated, the particular column is said to be "inactive" and it cannot be written to. For example, such a write enable may be provided in a system as illustrated in U.S. Pat. No. 4,253,102. In order to read information out of the memory cell, the user must be able to activate the memory cell. This can be done by providing the memory cell with a "read enable wire." If this read enable wire is activated, then the information stored in the memory cell is communicated out of the cell. Alternatively