Connectionism, or neuronlike computing, developed out of attempts to understand how the human brain works at the neural level and, in particular, how people learn and remember. In 1943 the neurophysiologist Warren McCulloch of the University of Illinois and the mathematician Walter Pitts of the University of Chicago published an influential treatise on neural nets and automatons, according to which each neuron in the brain is a simple digital processor and the brain as a whole is a form of computing machine. As McCulloch put it subsequently, “What we thought we were doing (and I think we succeeded fairly well) was treating the brain as a Turing machine.”

Creating an artificial neural network

It was not until 1954, however, that Belmont Farley and Wesley Clark of MIT succeeded in running the first artificial neural network—albeit limited by computer memory to no more than 128 neurons. They were able to train their networks to recognize simple patterns. In addition, they discovered that the random destruction of up to 10 percent of the neurons in a trained network did not affect the network's performance—a feature that is reminiscent of the brain's ability to tolerate limited damage inflicted by surgery, accident, or disease.

The simple neural network depicted in the figure illustrates the central ideas of connectionism. Four of the network's five neurons are for input, and the fifth—to which each of the others is connected—is for output. Each of the neurons is either firing (1) or not firing (0). Each connection leading to N, the output neuron, has a “weight.” What is called the total weighted input into N is calculated by adding up the weights of all the connections leading to N from neurons that are firing. For example, suppose that only two of the input neurons, X and Y, are firing. Since the weight of the connection from X to N is 1.5 and the weight of the connection from Y to N is 2, it follows that the total weighted input to N is 3.5. As shown in the figure, N has a firing threshold of 4. That is to say, if N's total weighted input equals or exceeds 4, then N fires; otherwise, N does not fire. So, for example, N does not fire if the only input neurons to fire are X and Y, but N does fire if X, Y, and Z all fire.

Training the network involves two steps. First, the external agent inputs a pattern and observes the behaviour of N. Second, the agent adjusts the connection weights in accordance with the rules:

1. If the actual output is 0 and the desired output is 1, increase by a small fixed amount the weight of each connection leading to N from neurons that are firing (thus making it more likely that N will fire the next time the network is given the same pattern);
   
   
2. If the actual output is 1 and the desired output is 0, decrease by that same small amount the weight of each connection leading to the output neuron from neurons that are firing (thus making it less likely that the output neuron will fire the next time the network is given that pattern as input).

The external agent—actually a computer program—goes through this two-step procedure with each pattern in a training sample, which is then repeated a number of times. During these many repetitions, a pattern of connection weights is forged that enables the network to respond correctly to each pattern. The striking thing is that the learning process is entirely mechanical and requires no human intervention or adjustment. The connection weights are increased or decreased automatically by a constant amount, and exactly the same learning procedure applies to different tasks.

(A mnemonic is) any device for aiding the memory. Named for Mnemosyne, the goddess of memory in Greek mythology, mnemonics are also called memoria technica (Latin: “memory technique”). The principle is to create in the mind an artificial structure that incorporates unfamiliar ideas or, especially, a series of dissociated ideas that by themselves are difficult to remember. Ideally, the structure is designed so that its parts are mutually suggestive.

Early methodologies

Numerous attempts have been made to invent mnemonic systems, or generalized codes to improve the all-around capacity to remember. The Greek and Roman system of mnemonics was founded on the use of mental places and signs or pictures in terms of the location of the items of interest. The method combines a familiar structure (locus) and the item or thing to be remembered (res). This mnemonic method is referred to as loci et res, or method of loci, and is an effective way to remember a series of items. The most usual method is to choose a large house in which the rooms, walls, windows, decorations, and furniture are severally associated with certain names, phrases, events, or ideas by means of symbolic pictures. To recall these items, it is necessary only to search the rooms of the house mentally until the particular place where the imagination deposited them is discovered. In accordance with this system, if it is desired to fix a historic date in the memory, it is localized in an imaginary town divided into a certain number of districts. Each district has 10 houses; each house has 10 rooms; and each room has 100 quadrates, or memory places. They are partly on the floor, partly on the four walls, and partly on the roof. By means of this system, the traditional date of the invention of printing in Europe (1440) could be fixed in the memory by mentally placing a book or some other symbol of printing in the 40th quadrate of the fourth room of the first house of the imaginary town.

Later developments

A related method, called linking or chaining, associates any pair of items—a pen and a chair, for example—and then links those items with a third, the chain proceeding indefinitely. Interaction, as opposed to mere association, is necessary—one could imagine the pen writing on the chair, for example—as one word aids in recalling the next. Associations may even link to tell a story. This method has proved effective in use with grammar-school children as well as with adults.

A similar technique is the peg-word system, a memory aid that involves linking words with numbers. It is utilized by creating mental associations between items to be remembered and items that are already associated with numbers (the latter is a relatively simple task, as the item-number pairs often rhyme). For example, to remember the seven deadly sins—lust, pride, greed, anger, sloth, envy, and gluttony—the number one could be associated with a bun, two with a shoe, three with a tree, four with a door, five with a hive, six with sticks, and seven with heaven. Then lust would be remembered by imagining a man drooling over a cinnamon bun, pride would be remembered by picturing a man polishing his expensive shoes, greed would be remembered by envisioning the word hanging from a tree in place of fruit, and so on.

A more common mnemonic device is rhyming. Grouping items in rhymed verse has long been a popular mnemonic technique, from the “gender rhymes” of the Latin grammars to the verse for remembering the number of days in the months of the year (“Thirty days hath September, April, June, and November…”). Other examples include “i before e, except after c” and “In 1492, Columbus sailed the ocean blue.”

Another method, chunking, involves grouping individual pieces of information in a manner that makes them easier to remember (i.e., relation, hierarchical importance, function, and so on). For example, the individual digits 1, 9, 6, and 1 may be easier to remember as the year 1961; the digits 6, 2, 5, 4, 3, 9, and 1 might be more readily recalled as the telephone number 625-4391; and a grocery list might be more easily remembered by food category (i.e., fruits, vegetables, and so on).

Acronyms and acrostics are also useful mnemonic devices. An acronym is a familiar or memorable word composed of the first letters of a series of words to be remembered. For example, a widely used acronym for remembering the five Great Lakes—Huron, Ontario, Michigan, Eerie, and Superior—is HOMES. An acrostic is a phrase or sentence in which the first letter of each word stands for the first letter in a list of words to be remembered. For instance, an acrostic frequently used to help mathematics students remember the order of operations is Please Excuse My Dear Aunt Sally (Parentheses, Exponents, Multiplication/Division, and Addition/Subtraction). Acronyms and acrostics can be particularly useful when items must be remembered in a specific order.

One point stressed by mnemonists is that bizarre images can amplify the effectiveness of a memory aid, a phenomenon known as the bizarreness effect. The bizarreness effect holds that items associated with bizarre sentences or phrases are more readily recalled than those associated with common sentences or phrases. For example, a student might better remember the five main vowels of the English alphabet (A, E, I, O, U) by visualizing the bizarre sentence An Elephant Is Orbiting Uranus as opposed to the common sentence An Explorer Is Orbiting Uranus.

Scientific interest in mnemonics was heightened in 1968 when the renowned Soviet neuropsychologist Aleksandr R. Luria suggested, in The Mind of a Mnemonist, that the field was worthy of deeper psychological study. Luria described a man with synesthesia— a neurological condition in which the stimulation of one the five senses results in the simultaneous stimulation of one or more of the remaining senses— who had a remarkable memory.

Adenosine triphosphate as the currency of energy exchange

When the terminal phosphate group is removed from ATP by hydrolysis, two negatively charged products are formed, ADP3- and HPO42- (a phosphate group).

These products are electrically more stable than the parent molecule and do not readily recombine. The total free energy (G) of the products is much less than that of ATP; hence energy is liberated (i.e., the reaction is exergonic). The amount of energy liberated under strictly defined conditions is called the standard free energy change (?G?); this value for the hydrolysis of ATP is relatively high, at -8 kilocalories per mole. (One kilocalorie is the amount of heat required to raise the temperature of 1,000 grams of water one degree centigrade.) Conversely, the formation of ATP from ADP and inorganic phosphate (Pi) is an energy-requiring (i.e., endergonic) reaction with a standard free energy change of +8 kilocalories per mole.

The hydrolysis of the remaining phosphate-to-phosphate bond of ADP is also accompanied by a liberation of free energy (the standard free energy change is -6.5 kilocalories per mole); AMP hydrolysis liberates less energy (the standard free energy change is -2.2 kilocalories per mole).

Source: "Metabolism." Encyclopædia Britannica Ultimate Reference Suite, 2013.

Qualitative tests to distinguish alternative theories

At the time that Augustin-Jean Fresnel presented his wave theory of light to the French Academy (1815), the leading physicists were adherents of Newton's corpuscular theory. It was pointed out by Siméon-Denis Poisson, as a fatal objection, that Fresnel's theory predicted a bright spot at the very centre of the shadow cast by a circular obstacle. When this was in fact observed by François Arago, Fresnel's theory was immediately accepted.

Another qualitative difference between the wave and corpuscular theories concerned the speed of light in a transparent medium. To explain the bending of light rays toward the normal to the surface when light entered the medium, the corpuscular theory demanded that light go faster while the wave theory required that it go slower. Jean-Bernard-Léon Foucault showed that the latter was correct (1850).

The three categories of experiments or observations discussed above are those that do not demand high-precision measurement.