Are we the sum of our memories? If so, given the fragile nature of Human memory—how easily memories can be lost, altered, or even created—what are we really? If that’s not angst-inducing enough, what are you, then, if you’re a clone made by the Tleilaxu, a ghola, and your memories—which aren’t really even your memories—are all implanted? Csilla Csori explores these very issues .
S INCE MANKIND FIRST SAT AROUND A FIRE telling scary stories, we’ve had tales of people coming back from the dead in one form or another. Modern tellings substitute science for the supernatural, but these stories continue to capture our imagination and leave us wondering, is it possible? Would such a person be the mindless zombie of countless horror movies, or a resurrected Lazarus?
These questions will not always be academic. People used to be considered dead when their hearts stopped, but not anymore. Every day, people whose hearts have stopped beating for brief periods of time—sometimes their hearts are even stopped on purpose during surgery—are brought back to life, and they are essentially the same person as before. The next logical step is to consider what would happen if we could overcome brain death. Would the revived person be the same, have the same personality, and retain all of their memories?
In Dune Messiah , we are first introduced to the ghola as a body resurrected through medical means. Our ghola, Duncan Idaho, has a consciousness of his own but no memory of his former life or his death. He is most definitely not a mindless zombie, and, in fact, is resurrected with the capability to learn new skills which he didn’t possess in his previous life. He is a person, but is he the same person, the same Idaho who lived and died before? Not at first. Although certain voices and places seem familiar, he doesn’t regain any actual memories until one traumatic event unlocks the past, and all of the memories from his former life come flooding back at once. If the brain is like a computer, then it is as if his memories are stored in a hidden file system to which he does not have access. Once he acquires the key to unlocking that system, all of the files (or memories) within are opened to him at once, and he knows himself as Idaho.
The news is full of stories in which authorities confiscate a suspect’s computer and recover a hoard of incriminating files which the suspect had deleted. If a computer can retain deleted files, what about a human brain? Should this analogy of recovering information from a damaged brain give us hope for the Terri Schiavos of the world? If we could repair and regrow brain cells, would her personality still be in there somewhere, fully intact and just needing the right key to unlock it? To answer these questions, we need to examine how a brain stores and retrieves memories, and how this process compares with computer memory.
Like a computer, your brain has storage systems for short-term and long-term memory, and a central processing unit, known as the hippocampus , which connects the two. Your hippocampus filters data—determining what is relevant—from short-term memory to long-term memory. However, the manner in which data is stored and then accessed later is different, and that is where the key to recovering lost memories lies.
Most computer users are familiar with hierarchical file systems, which are basically made up of a main directory (or folder) containing files and subdirectories. The subdirectories can, in turn, contain both files and additional subdirectories. Users navigate up or down the hierarchical structure to locate a specific file in a specific directory. You might expect that all of the files in a particular directory are stored next to each other in the computer’s memory, but this is not always the case. In fact, a single file may be broken into fragments and stored in several locations. This fragmentation occurs when files are edited and increase or decrease in size. On a brand-new disk (or other chunk of storage), the computer’s operating system starts at one part, writing data in an orderly fashion, and if the data never changes, it continues until the disk is full. But data files are almost never static; users are constantly adding on to files, deleting entire files, and otherwise changing the amount of memory needed to store a subdirectory or a particular file. As data is deleted, chunks of memory become available for new information, making holes in the nice, orderly system. When a file increases in size, if there is not enough memory in the original location to store the entire file, then the computer will look for an additional chunk of memory to store the second piece of the file. This process can be repeated many times, and a single file may end up stored as several pieces spread out over the disk.
The exact method a computer uses to keep track of all the pieces differs between operating systems, but it basically uses some kind of master reference table. When a user deletes a file, the actual data is not erased—only the entry in the reference table gets deleted. This tells the computer that the particular chunk of memory on which that file is stored is now available for writing new data. But the old data will sit there until it is overwritten, so that is why it is possible to recover deleted files from a computer.
Does the brain work in an analogous way, allowing us to recover lost memories? Your brain also stores pieces of memories in different locations; but, unlike a computer, it does not store information sequentially. Different types of sensory signals, such as sight, sound, and taste, are processed in different regions of your cortex and routed to your hippocampus. After filtering, the hippocampus sends these bits of information back to their respective regions and creates neural links between them. These links are strengthened by repetition (for example, by repeating a list) and by emotional factors such as the personal relevance of the information. Your hippocampus keeps track of all of the links and associations, indexing and cross-linking with similar information. It seems similar to the master reference table in a computer operating system, but it is much more complex. Even though a computer may break a file into several pieces for storage, it still considers a file to be one discrete unit. The computer has no way of examining file content and determining that the letter you wrote to Grandma last week is in any way connected with the photo of her hugging you as a child. Your brain’s reference system, on the other hand, is able to cross-link information from memories that are widely separated in time and location, and makes connections based on everything from strong emotions to mundane details.
This interweaving of memories strengthens associations, but it can also muddle memory recall and make it unreliable. When you recall the memory of an event, you are not opening a single file containing all of the data. Rather, you are dynamically reconstructing the memory from its various components. The process is associative, so one thing, like a particular song or smell, can trigger an associated piece of the memory, which triggers another, and so on. The ease and accuracy of your recall depends on the number and strength of the neural links, which, in turn, are dependent on such factors as how long ago the event occurred, when you last remembered it, and whether it is similar to other events in your memory. In the process, pieces of memories can get confused and mixed in with one another. For example, a married couple who has had several arguments over money may mix up what was said during which argument when trying to recall one particular confrontation. If they later have to testify in court as to what was said, they may give different accounts and yet each will believe they are telling the truth. In addition, the process of memory reconstruction is further clouded by current emotions and motivations. So, unlike a computer file, which is the same each time you open it, your memory of an event will differ at different times in your life.
Consider again our ghola, who has no memory of his former life. If those memories are still stored in his brain, how might they be accessed? Amnesia is often temporary, with people gradually recalling some or all of their missing memories. Our ghola’s brain has been repaired, so there is no physical damage preventing access. If the neural links are intact, then it should be as simple as placing him in an environment which will trigger the old memories. It is unlikely that everything would return at once. A familiar face or voice would bring back a flood of associations, and, over time, the entirety of his memories should return. Of course, he is not exactly the same person, especially after the trauma of remembering his own death—but he is, for all intents and purposes, Idaho.
However, it’s not that easy, because our ghola’s memories are locked away in that hidden file system. In searching for a physical cause for the block, you might suppose that there is something in his hippocampus, or CPU, that is preventing access, but it is not that straightforward. Once the associations between neurons—the neural links—reach a certain strength, they become independent of the hippocampus, and the neurons can trigger each other directly. So his oldest, strongest, and most well-connected memories are not controlled by the hippocampus at all. In fact, damage to the hippocampus has the opposite effect on memory than what our ghola is experiencing. Rather than causing retrograde amnesia—the inability to recall past events—a damaged hippocampus causes anterograde amnesia—the inability to acquire new memories. Without the hippocampus, short-term memories can never be translated into long-term memories, and they are lost forever. Drew Barrymore’s character, Lucy, in 50 First Dates and Guy Pearce’s character, Leonard, in Memento are two examples of people suffering from anterograde amnesia.
Therefore, there is no simple physical explanation for a total memory block in the presence of familiar surroundings. Due to the distributed, associative nature of memory, there is no central switch to turn on and off, no single access point which can be hidden or encrypted. Even in cases where a person suffers from severe retrograde amnesia due to lesions on the brain, such as in Alzheimer’s disease, early childhood memories generally remain intact.
Perhaps our ghola’s memory loss is not due to a physical cause, but a psychological one. The trauma of dying is surely something he would want to block out. Although rare, there have been cases where people suffered from amnesia after being the victim of a violent crime, but the amnesia was associated with a confused state and only lasted a short time. What remains, then, is the complex and controversial subject of repressed memories, a concept which is often associated with childhood abuse. Can a memory be forgotten, either intentionally or subconsciously, and then be remembered later? According to the American Psychological Association, both phenomena do occur, but the mechanism is not well understood. The accuracy of recovered memories is questionable; as the brain reconstructs those memories from their component parts, the person’s emotions and intent influence the result. Memories are not perfect recordings of events, but rather, impressions colored by our emotional state both at the time the memory was formed and at the time it is remembered. To further confuse matters, it is possible to construct false memories of events that never occurred.
Even though the concept of repressed memory is possible, it does not offer a satisfying explanation for the total amnesia our ghola is experiencing. In recorded cases of repressed and recovered memories, the phenomenon was localized to those memories associated with the traumatic event. Our ghola might not remember the circumstances of his death, but he would not suppress the memories of his entire life. So a psychological cause for his type of memory loss is no more likely than a physical one.
We have looked at the question of access, of how memories are recalled, and whether they could be hidden from the conscious mind until triggered by a single event or whether memories would return in bits and pieces over time. But this assumes that the intact memories are in the brain to begin with. The next question is of storage, of whether old memories would remain in the brain at all. The answer depends on the type of ghola, since there are two distinct methods for creating them.
In the time when Dune Messiah is set, the process of creating a ghola requires the entire body of the original person. A ghola is literally a corpse brought back to life. The dead flesh of Idaho is placed in a tank where his damaged tissue is repaired, and a person emerges, alive and conscious. This person has no memory of his past, but since he has the same brain, he still has the neural connections (the file system of memories) created by all of the events of his life. Time is the greatest limiting factor, since neural links weaken with disuse. If a lengthy period passes before our ghola is exposed to memory triggers, some of his past may be lost. But his oldest and strongest memories will remain for a long time, and chances are good that he will regain at least some of his former identity.
However, by the time God Emperor of Dune takes place, technological advances have changed the process dramatically. The gholas of Idaho are not the same body repaired and resurrected again and again. They are grown from mere cells of the original person, and there can be more than one of them alive at any given time. In other words, they are clones. This has completely different implications for the possibility of memory retention, because it requires the transfer of memories from one body to another. Like any clone, the adult gholas of Idaho are created using DNA as the means of coding information into the copy. What we know about the way memories are stored and retrieved in the brain involves neurological and chemical processes. There is no research to indicate that DNA stores specific memories, such as the events in a person’s life. As our ghola grows in his tank, his DNA dictates the basic structure of his brain, but it does not stimulate the neural links which are key to the creation of memories. When he emerges, even though he is physically an adult, he is essentially a newborn person. Unfortunately, our Idaho clone has no inherent memories of the original Idaho’s life.
What about transferring memories, downloading them from the original into a copy? Preserving the original brain indefinitely poses a problem, so it is more practical to download memories into a permanent storage system, such as a computer disk or flash drive, and upload the information into our ghola as needed. This system requires a working interface between the computer, the hippocampus, and other parts of the brain; but once that is achieved, it is a matter of sending signals through the brain and recording the position and strength of electrical impulses. This gives us a snapshot of the physical structure of Idaho’s brain at the time of his death.
If we re-create this physical structure in our ghola’s brain, is it the equivalent of uploading Idaho’s memories? More than a question of physical and biochemical requirements and limitations, the heart of this query asks what makes us who we are. If we can create physical clones of Idaho and give them all the same memories, experiences, and personality, then what makes any of them a unique individual? If the clones are perfect recreations, do terms like “original” and “copy” even have any meaning? The conclusion of these questions may have to wait until the first ghola emerges from his tank and speaks the answer.
Until cloning reaches that level of technology, our first type of ghola—the resurrected person—is the kind we will have to deal with. It is not just a subject for speculative fiction, but a topic for present-day discussion. As medical science advances, the moment when a person is beyond resuscitation gets pushed further and further back. Like Miracle Max in The Princess Bride , our doctors can determine if a person is just “mostly dead,” and therefore partly alive. Machines can assist the heart and lungs to function until the body heals sufficiently to work on its own. Unfortunately, brain science has not yet advanced to the point where we can repair damaged brain cells, but that, too, is in our near future.
What will a real-life ghola, a person returned from brain death, be like? Will he remember any of his past, or will he be an entirely different person? In addition to impaired function, people who suffer from non-lethal brain damage often experience memory loss and even changes in personality. Repairing damaged cells would clearly return them to normal functioning, but what about memories? A cell which sustained only partial damage would retain some of its neural connections. A newly grown brain cell would not, but if it were connected to undamaged cells, the links from those healthy cells might be sufficient for the memory connection. Memory recovery would depend greatly on the extent of initial damage, but the distributed nature of memory works to our benefit here, as it’s unlikely that all areas associated with any particular memory would have been damaged.
When medical technology provides us with a method for repairing and regrowing brain cells, the diagnosis of brain death may cease to exist. Just as people who suffer cardiac arrest today can have their hearts restarted, people who suffer severe brain damage may someday have their brains jump-started, or otherwise brought back online. For the person returned to life in this manner—our real-life ghola—this means that he has a chance of regaining at least parts of his memories, especially if he is in familiar surroundings which will trigger memory associations. And that’s really all our ghola needs: a chance of recovery, the hope that his memories may trickle back and that he will regain some semblance of the person he was before.
CSILLA CSORI is a programmer/analyst at the San Diego Supercomputer Center. She works primarily on database and software development for business applications, and she also moonlights as a gremlin hunter for her colleagues when their computer programs start acting funny. Recently, she released version 5.1 of ProBook grant application software she authored for the University of California. It’s one of those pesky projects that started small but took on a life of its own, and now, like Doctor Who ’s Cybermen, keeps coming back to demand more upgrades. She gained an interest in brain function in college, where she earned extra cash by volunteering for cognitive experiments at the National Institutes of Health. In her spare time, she enjoys playing softball, kayaking, and any other excuse to be outdoors in San Diego’s perfect weather.
American Psychological Association. “Questions and Answers about Memories and Childhood Abuse.” Learning and Memory . Aug. 1995 http://www.apa.org/topics/memories.html
Dubuc, Bruno. “Memory and the Brain.” The Brain From Top to Bottom . http://thebrain.mcgill.ca/